450
Denna publikation säljs av
This publication is available from
VTT INFORMATIONSTJÄNST
PB 2000
02044 VTT
Tel. (09) 456 4404
Fax (09) 456 4374
VTT INFORMATION SERVICE
P.O.Box 2000
FIN–02044 VTT, Finland
Phone internat. + 358 9 456 4404
Fax + 358 9 456 4374
ISBN 951–38–5878–2
ISSN 1235–0621
A guide to physical property characterisation of biomass-derived fast pyrolysis liquids
Tätä julkaisua myy
VTT TIETOPALVELU
PL 2000
02044 VTT
Puh. (09) 456 4404
Faksi (09) 456 4374
VTT PUBLICATIONS 450
A guide to physical property characterisation of biomass-derived fast
pyrolysis liquids is a revised and updated version of VTT Publication 306:
Physical characterisation of biomass-based pyrolysis liquids. Application of
standard fuel oil analyses by Oasmaa et al. published in 1997. The objective
of this work was to test the standard methods developed for characterization
of petroleum fuels on biomass pyrolysis liquids and modify them if necessary
to accommodate the significant differences in properties between these two
groups of fuels. Compared to the previous publication, this work extends the
method validation tests on more biomass-derived liquids and more physical
properties. It includes the recommendations on testing procedures for twophase liquids obtained from pyrolysis of forest residues, the most likely
feedstock for producing pyrolysis liquids. A chapter on transporting and
handling pyrolysis liquids that considers the health and safety aspects of the
operations is included. The publication is mostly focused on practical issues
related to pyrolysis liquid characterization such as sampling techniques and
recipes for the tests. Summarizing, A guide to physical property
characterisation of biomass-derived fast pyrolysis liquids will serve as a
guide and a reference for the researchers and technicians working with
biomass pyrolysis liquids.
V T T
P U B L I C A T I O N S
Anja Oasmaa & Cordner Peacocke
A guide to physical property
characterisation of biomassderived fast pyrolysis liquids
V
T
T
P
u
b
l
i
c
a
V
T
T
P
u
b
l
i
c
a
V
T
T
P
u
b
l
i
c
a
V
T
T
P
u
b
l
i
c
a
V
T
T
P
u
b
l
i
c
a
V
T
T
P
u
b
l
i
c
V
T
T
P
u
b
l
i
V
T
T
P
u
b
l
V
T
T
P
u
b
l
P
t
i
o
n
s
t
i
o
n
s
t
i
o
n
s
t
i
o
n
s
t
i
o
n
s
a
t
i
o
n
s
c
a
t
i
o
n
s
i
c
a
t
i
o
n
s
i
c
a
t
i
o
n
s
TECHNICAL RESEARCH CENTRE OF FINLAND
ESPOO 2001
VTT PUBLICATIONS 450
A guide to physical property
characterisation of biomass-derived
fast pyrolysis liquids
Anja Oasmaa
VTT Energy
Cordner Peacocke
Conversion And Resource Evaluation Ltd.
TECHNICAL RESEARCH CENTRE OF FINLAND
ESPOO 2001
ISBN 951–38–5878–2 (soft back ed.)
ISSN 1235–0621 (soft back ed.)
ISBN 951–38–6365–4 (URL:http://www.vtt.fi/inf/pdf)
ISSN 1455–0849 (URL:http://www.vtt.fi/inf/pdf)
Copyright © Valtion teknillinen tutkimuskeskus (VTT) 2001
JULKAISIJA – UTGIVARE – PUBLISHER
Valtion teknillinen tutkimuskeskus (VTT), Vuorimiehentie 5, PL 2000, 02044 VTT
puh. vaihde (09) 4561, faksi (09) 456 4374
Statens tekniska forskningscentral (VTT), Bergsmansvägen 5, PB 2000, 02044 VTT
tel. växel (09) 4561, fax (09) 456 4374
Technical Research Centre of Finland (VTT), Vuorimiehentie 5, P.O.Box 2000, FIN–02044 VTT, Finland
phone internat. + 358 9 4561, fax + 358 9 456 4374
VTT Energia, Uudet energiatekniikat, Biologinkuja 3–5, PL 1601, 02044 VTT
puh. vaihde (09) 4561, faksi (09) 460 493
VTT Energi, Nya energiteknologier, Biologgränden 3–5, PB 1601, 02044 VTT
tel. växel (09) 4561, fax (09) 460 493
VTT Energy, New Energy Technologies, Biologinkuja 3–5, P.O.Box 1601, FIN–02044 VTT, Finland
phone internat. + 358 9 4561, fax + 358 9 460 493
Technical editing Maini Manninen
Text preparing Tarja Haapalainen
Otamedia Oy, Espoo 2001
Oasmaa, Anja & Peacocke, Cordner. A guide to physical property characterisation of biomassderived fast pyrolysis liquids. Espoo 2001. Technical Research Centre of Finland, VTT Publications 450. 65 p. + app. 34 p.
Keywords
pyrolysis liquid, biomass, fast pyrolysis, physical properties, fuel oil properties,
test methods, characterization, homogeneity, sampling, specifications
Abstract
This publication is a revised and updated version of VTT Publication 306:
Physical characterisation of biomass-based pyrolysis liquids, issued in 1997. The
main purpose of the on-going study is to test the applicability of standard fuel oil
methods developed for petroleum-based fuels to biomass-derived fast pyrolysis
liquids. New methods have also been tested and further developed to extend the
range of properties that may be accurately determined. The methods were tested
for pyrolysis liquids derived from hardwood, softwood, forest residue and straw.
Recommendations on transport, liquid handling and analyses are presented. In
general, most of the standard methods for fuel oils can be used as such, but the
accuracy of the analyses can be improved by minor modifications. Homogeneity
of the liquids is a critical factor in the accurate analyses, and hence procedures for
its verification are presented.
3
Preface
VTT (Finland) has had a central role in analysing both liquid and solid fuels for
the Finnish industry for over two decades. Considering these experiences, the
standard fuel oil analyses developed for petroleum-based fuels were applied to
pyrolysis liquids. This original study was published as VTT Publication 306
(http://www.inf.vtt.fi/pdf/publications/1997/P306.pdf) on physical characterisation of biomass-based pyrolysis liquids, issued in 1997. In this publication, the
former study has been expanded to include more feedstocks, analyses and developments in test methods from other sources to encompass as many properties of
the liquids as possible. Additional information on health and safety are also
summarised. At Aston University, special analyses, e.g. specific heat capacity
and conductivity, have been carried out by C. Peacocke, who presently works in
a consulting company Conversion And Resource Evaluation Ltd. (CARE). This
company is a small to medium size enterprise operating since 1996 in the thermal conversion of biomass and wastes for energy, chemicals and speciality derived products. The focus of CARE's work has been on small-scale gasification,
fast pyrolysis, and the techno-economic assessment of such systems for fuels
and chemicals. In this revised report, the test results are summarised and instructions for liquid handling are suggested.
Special thanks are due to Stefan Czernik (NREL, USA) for the review of the
publication and valuable comments. The authors wish to acknowledge Dietrich
Meier (IWC, Germany), Tony Bridgwater (Aston University, UK) and other
members of the Pyrolysis Network for Europe (PyNe) for their valuable conribution and comments during the work. The valuable discussions with Steven Gust
(Fortum, Finland) are greatly acknowledged. The VTT project was funded by
the National Technology Agency (Tekes). Analytical work at VTT was carried
out by Eija Tapola, Raili Silvasti, Jaana Korhonen, Eeva Kuoppala, Päivi Koponen, Paula Käyhkö, Sirkka-Liisa Huru, and Kaisa Lanttola. Maija Korhonen's
contribution is warmly acknowledged for editing the report.
Espoo, December 2001
Authors
4
Contents
Abstract................................................................................................................. 3
Preface .................................................................................................................. 4
1. Introduction..................................................................................................... 7
2. Pyrolysis liquids production ......................................................................... 11
3. Health, safety and transport .......................................................................... 14
3.1 Toxicity of pyrolysis liquids and their derivatives ............................ 14
3.2 Material and Safety Data Sheets (MSDS) ......................................... 15
3.3 Transport, storage and handling ........................................................ 15
4. Homogeneity and sampling .......................................................................... 18
4.1 Homogeneity of pyrolysis liquids ..................................................... 18
4.2 Homogenisation and sampling .......................................................... 19
5. Composition.................................................................................................. 20
5.1 Water ................................................................................................. 20
5.2 Solids ................................................................................................. 22
5.3 Particle size distribution .................................................................... 23
5.4 Elemental composition ...................................................................... 25
5.5 Ash and metals .................................................................................. 25
6. Fuel oil properties ......................................................................................... 28
6.1 Density............................................................................................... 28
6.2 Viscosity............................................................................................ 29
6.3 Pour point .......................................................................................... 33
6.4 Cloud point ........................................................................................ 34
6.5 Conradson carbon residue ................................................................. 34
6.6 Heating value..................................................................................... 36
6.7 Flash point ......................................................................................... 37
6.8 Ignition temperature .......................................................................... 38
6.9 Thermal conductivity, specific heat capacity and electrical
conductivity ....................................................................................... 39
5
6.10 Lubricity ............................................................................................ 40
7. Unique properties of pyrolysis liquids.......................................................... 42
7.1 Acidity and material corrosion .......................................................... 42
7.2 Solubility ........................................................................................... 44
7.2.1 Water solubility ............................................................................ 44
7.2.2 Solubility in other solvents........................................................... 46
7.3 Stability ............................................................................................. 47
8. Fuel oil specifications ................................................................................... 51
9. Recommendations......................................................................................... 54
References........................................................................................................... 58
Appendices A–H
6
1. Introduction
Biomass pyrolysis liquids differ significantly from petroleum-based fuels in both
physical properties and chemical composition. These liquids are typically high in
water and solids, acidic, have a heating value of about half of that of mineral
oils, and are unstable when heated, especially in air. Pyrolysis liquids are highly
polar containing about 50 wt% oxygen (~ 40 wt% of dry matter), while mineral
oils contain oxygen in ppm levels. Pyrolysis liquids are therefore highly oxygenated and are unlike conventional mineral oils. The unusual properties of the liquids must therefore be taken into careful consideration in the determination of
fuel oil qualities and chemical applications for the liquids. For fuel applications,
the present consensus from end users on significant liquid properties for combustion applications are stability, homogeneity, water lower heating value, viscosity, liquid density, solids content and to a limited extent chemical composition. Depending on the physical property values, the suitability of the liquid for
use in a boiler, engine or turbine application can be assessed. For non-fuel applications, other properties may of interest [number of phases, pH, detailed chemical composition]. Due to these differences, the standard fuel oil methods developed for mineral oils are not always suitable as such for pyrolysis liquids. Elliott
[1983] and McKinley et al. [1994] have previously addressed the importance of
the reliability of the analytical methods. The typical properties of pyrolysis liquids from different feedstocks are given in Table 1. [Agblevor et al. 1996, Sipilä
et al. 1998, Radovanovic et al. 2000].
Research on analysing physical properties of pyrolysis liquids has been carried
out since the 1980s at PNL, USA [Elliott 1983], NREL (formerly SERI), USA
[Chum & McKinley 1988, Milne et al. 1990, Czernik et al. 1994], and at the
former B.C.Research, Canada [McKinley 1989, McKinley et al. 1994]. Rick and
Vix [1991] published one of the few reviews on product standards for pyrolysis
liquids. Fagernäs [1995], Oasmaa et al. [1997], Meier et al. [1997] and Oasmaa
& Meier [Brigdwater et al. 1999 and 2001] have published up-to-date data on
physical properties of pyrolysis liquids. Elliott [1983] and Diebold et al. [1997]
have proposed fuel oil specifications for pyrolysis liquids.
Several round robins have been carried out in order to assess the relevant analytical procedures for pyrolysis liquids's analyses. The first IEA (International
Energy Agency) thermochemical round robin was organised in 1988 as part of
7
the IEA Voluntary Standards Activity led by B.C. Research [McKinley et al.
1994]. Two separate round robins were initiated in 1997: one within EU PyNe
(Pyrolysis Network) and the other within IEA PYRA (Pyrolysis Activity). The
objective of the EU PyNe round robin was to compare existing analytical methods without any restrictions. The main objective of the IEA PYRA round robin
was to determine the interlaboratory precision and methods applied for elemental composition, water, pyrolytic lignin and main compounds. The latest round
robin was carried out by 12 laboratories during January - March 2000. The aim
was to compare the analyses, not the pyrolysis liquids. Instructions for analyses
were given. [Bridgwater et al. 2001]
Based on these round robins following conclusions were drawn: liquid sample
handling plays a very important role, the precision of carbon and hydrogen is
very good, oxygen by difference is variable and oxygen by direct determination
is poor, water by Karl-Fischer titration is accurate but should be calibrated by
water addition method, accuracy of density is good, high variations were obtained for nitrogen, viscosity, pH, and solids, the methods for the determination
of pyrolytic lignin and measuring stability of the liquid should be improved,
[Bridgwater et al. 2001]
This publication is an updated version of a study on testing and modifying standard fuel oil analyses [Oasmaa et al. 1997]. Additional data has been included to
address the wide spectrum of properties that may be required in different applications and to assist in the design of process equipment and power generation
systems.
8
Table 1. Pyrolysis liquids properties from different feedstocks.
Source
Physical
property
Feedstock
Ensyn
Mixed
hardwood
22.0
2.5
NREL (hotgas filtered)
Poplar
18.9
Water, wt%
2.8
pH
Density (at
1.20
1.18
15°C), kg/dm3
Elementals
(dry), wt%
57.3
56.4
- C
6.3
6.2
- H
36.2
37.1
- O (by diff.)
0.18
0.2
- N
0.02
<0.01
- S
<0.01
0.01
- Ash
10
460
- K+Na, ppm
8
3
- Cl, ppm
18.7
17.0
HHV, MJ/kg
HHV (dry),
22.3
23.1
MJ/kg
Viscosity, cSt
n.d.
128
- 20°C
50
13.5
- 50°C
55
64
Flash point, °C
-36
-25
Pour point, °C
Solubility, wt%
insolubles in
0.045
0.045
- ethanol
- methanoln.d.
n.d.
dichloromethane
*Bottom phase (90 wt% of total liquid)
n.d. = not determined
NREL
VTT
VTT
VTT
Switch
grass
n.d.
n.d.
Pine
Straw
16.6
2.6
19.9
3.7
Forest
residue*
24.1
2.9
n.d
1.24
1.19
1.22
55.8
6.9
36.3
0.79
0.03
n.d.
128
1 900
55.8
5.8
38.2
0.1
0.02
0.03
20
30
19.1
55.3
6.6
37.7
0.4
0.05
0.14
2
330
18.5
56.6
6.2
36.9
0.1
0.03
0.08
60
<100
17.4
23.8
22.9
23.1
23.0
n.d.
n.d.
n.d.
n.d.
n.d.
31
n.d.
-19
55
11
56
-36
152
29 (40°C)
42
-12
n.d.
0.30
0.3
0.10
n.d.
n.d.
n.d.
0.02
Recent data on analytical development, stability, transport, storage, handling and
health and safety issues are included. The aim of this publication is to provide
guidelines for pyrolysis liquid producers and end-users in determining the fuel
oil quality of pyrolysis liquids. A further objective is to provide a baseline for
standardising the most important analytical methods of pyrolysis liquids.
9
As pyrolysis liquids will be pumped, shipped, atomised and combusted, as wide
a range of measured properties, mainly from VTT, but other sources have been
included to allow this publication to be used as a guide to test methods, a reference for measured values for comparison and as a guide to the determination of
properties primarily for fuel applications.
A brief summary of technology developers and providers is given (chapter 2)
providing details up to August 2001. Chapter 3 allows the user to assess the hazard posed by the liquids for transport, storage and handling and their toxicity
with some indications of precautions to be taken in use. Chapter 4 then addresses
how samples may be assessed for homogeneity, prior to use. Chapter 5 summarises the methods for determination of basic liquid properties [water content,
solids and particle size, elemental composition, ash and metals content] to determine initial suitability for fuel use. Once the basic properties of the liquid
have been determined, principal fuel properties are reviewed [density, viscosity,
heating value, flash point, pour point and Conradson carbon residue]. Other useful properties for process design are included [thermal conductivity, specific heat
capacity and electrical conductivity and lubricity]. The more unusual properties
of pyrolysis liquids are summarised in Chapter 7 and proposed fuel oil specifications for different qualities of pyrolysis liquids are in Chapter 8. Conclusions and
recommendations on the fuel oil test methods and values for pyrolysis liquids
are presented in Chapter 9. More specific information on test methods and reproducibility are presented in Appendices A–H.
All comments from oil end-users or producers concerning the significance of
some specific methods or suggestions for further developments are highly appreciated ([email protected]).
10
2. Pyrolysis liquids production
Pyrolysis liquids can be produced with a range of processes, each with their own
particular features. Our aim is not to review each type of process available, but
to give some indication of the degree of development of the technology.
A wide variety of configurations has been tested (Table 1), showing considerable diversity and innovation in meeting the basic requirements of fast pyrolysis,
and these have been extensively reviewed [Bridgwater & Peacocke 2000]. Most
fast pyrolysis processes give 65–75% liquids based on dry wood input.
The essential features of a fast pyrolysis reactor are very high biomass heating
and heat transfer rates; moderate and carefully controlled vapour temperature;
and rapid cooling or quenching of the pyrolysis vapours [Bridgwater 1995, Diebold & Bridgwater 1997]. Commercial operation is currently only been achieved
from a transport or circulating fluid-bed system, and only for food and flavouring products. Fluid-beds have also been studied extensively, are an ideal R&D
tool and have been scaled up to a pilot plant size with plans in hand for demonstration processes in several locations within the EU.
Most pyrolysis systems employ cyclones to remove char and ash from hot product gases and vapours. Some fine char is inevitably carried over from cyclones.
Unless removed by a hot vapour filter, which is still under development, char
will collect in the liquid and may be removed by liquid filtration employing, for
example, cartridge or rotary filters. Almost all of the biomass ash is retained in
the char, and hence, successful char removal generally involves successful ash
removal. Char separation, however, is difficult and may not be necessary in all
applications, i.e., chemical applications.
Char contributes to secondary cracking by catalysing secondary cracking in the
vapour phase. Rapid and complete char separation is therefore desirable. Even
char in the cooled collected liquid product contributes to instability problems,
accelerating slow polymerisation processes, which are manifested as increasing
viscosity.
The time and temperature profile between the formation of pyrolysis vapours
and their quenching to liquid affects the composition and quality of the liquid
product. High temperatures continue to crack the vapours and the longer the
11
vapours are at higher temperatures, the greater the extent of cracking. Although
secondary reactions become slow below around 350°C, some secondary reactions continue down to room temperature in the liquids, which contributes to the
instability of the pyrolysis liquid. Char also contributes to vapour cracking as
described above. A vapour residence time of a few hundred milliseconds is apparently necessary for optimum yields of certain chemicals, while fuels can tolerate a vapour residence time of up to around 2 seconds at 400 oC. A longer
residence time results in significant reductions in organics yields from cracking
reactions.
The collection of liquids has long been a major difficulty in the operation of fast
pyrolysis processes due to the nature of the liquid product that is mostly in the
form of aerosols rather than a true vapour. Quenching, i.e., the contact with a
cooled liquid is effective, but careful design and temperature control are needed
to avoid blockage from differential condensation. High-volatility components
are important in reducing liquid viscosity and therefore need to be recovered.
Electrostatic precipitation has been shown to be very effective in recovering
aerosols. In fluid-bed type systems, the vapour/aerosol concentration can be very
low, further increasing the difficulty of product separation due to the low vapour
pressure, significantly increasing the size of product recovery equipment. The
pyrolysis processes in operation, commissioning or under design (June 2001),
are listed with capacities and applications in Table 2.
12
Table 2. Pyrolysis liquids production processes, 2001 [> 10 kg/h].
Host organization
Country
Technology
kg/h
BTG
Netherlands
Rotating cone
2000
BTG/KARA
Netherlands
Rotating cone
250
Fuel and
chemicals
Operational
Dynamotive
Canada
Fluid bed
400
Fuel and
chemicals
Operational
Dynamotive
Canada
Fluid bed
80
Fuel and
chemicals
Operational
ENEL/Ensyn
Italy
Transported
bed
625
Fuel and
chemicals
Operational
Fortum
Finland
Own
500–
600
Fuel
Construction
Pyrovac
Canada
Vacuum
moving bed
3500
Fuel and
chemicals
Operational
Red Arrow/Ensyn
USA
Circulating
1250
transported bed
Fuel and
chemicals
Operational
Red Arrow/Ensyn
Canada
Circulating
125
transported bed
Fuel and
chemicals
Operational
RTI
Canada
Circulating
20
transported bed
Fuel and
chemicals
Operational
University of
Hamburg
Germany
Fluid bed
Waste
disposal
Fuel and
chemicals
Operational
50
Circulating
50
transported bed
Applications
Fuel and
chemicals
Status
Planned 2002
Operational
University of Laval Canada
Fluid bed
20
Fuel
Operational
VTT / Ensyn
Finland
20
Circulating
transported bed
Fuel
Operational
Wellman Proc.
Eng Ltd
UK
Fluid bed
Fuel
Commissioning
13
250
3. Health, safety and transport
3.1 Toxicity of pyrolysis liquids and their derivatives
In general, all toxicity testing methods can be divided into two categories. The
first category consists of tests that are designed to evaluate the overall effects of
compounds on experimental animals. The individual tests in this category differ
from each other concerning the duration of the test and the extent to which the
animals are evaluated critically for general toxicity. The tests are defined as
acute, prolonged and chronic toxicity tests.
The second category of tests consists of those tests that are designed to evaluate
in detail specific types of toxicity. The prolonged and chronic tests do not detect
all forms of toxicity but they may reveal some of the specific toxicities and indicate the need for more detailed studies. The second category of tests has been
developed to meet these needs. Examples of specific toxicity tests are:
Teratogenic
effects on the foetus
Reproduction tests
effects on reproduction
Mutagenic tests
effects on the genetic code system
Tumourigenicity
ability of agents to produce tumours [also known as
carcinogenicity test]
Neurotoxicity
effects on various behaviour patterns [also known as
behavioural tests]
Immunotoxicity
effects on the immune system
Other tests include skin and eye tests which are important for personal safety in
transport, storage and handling. There is no data on the carcinogenicity, teratoxicology or mutagenicity of pyrolysis liquids. This will become an area where
information will be required for the transportation of the liquids and assessing
the allowable exposure limits for those working with the liquids, whether at research scale or commercial production scale.
The health risks posed by fast pyrolysis liquids have not been clearly elucidated
so far, although recent reports [Diebold 1999] would suggest that there is a lim-
14
ited degree of mutagenicity and teratogenicity from the liquids, depending on
source, chemical composition and dosage.
Analogous work on liquid smoke [Putnam et al. 1999] suggests that depending
on the type of pyrolysis product, some cytotoxic and mutagenic effects may be
observed. Other studies suggest that aerosols of wood smoke are mutagenic
[Lewis et al. 1988]. Exposure to the aerosols is to be avoided. Skin contact is
also to be prevented by the use of protective gloves, clothing and safety glasses.
3.2 Material and Safety Data Sheets (MSDS)
During the testing and utilisation of pyrolysis liquids, the liquids must be handled, stored, transferred and sampled. Material and Safety Data Sheets are essential for the safe use, handling, storage and transportation of fast pyrolysis liquids.
So far, there is no accepted standard, due to the high variability in properties and
limited developments in commercial systems. Large pyrolysis liquid producers
(Ensyn, Dynamotive, BTG, Fortum) have their own MSDS for pyrolysis liquids.
Czernik (Bridgwater et al. 1999) prepared with IEA Bioenergy Pyrolysis Activity Group 1999 a MSDS based on all available data. A MSDS prepared at Aston
University (UK) which is suitable for inclusion with samples for shipment is
presented in Appendix A. All use of pyrolysis liquids requires the use of adequate safety equipment and facilities [protective clothing, chemical resistant
gloves, safety glasses or goggles and a well ventilated environment].
3.3 Transport, storage and handling
As the demand for fast pyrolysis liquids increases, it is important that they are
transported in a safe and environmentally secure manner [Peacocke & Bridgwater 2001]. The appropriate national and international regulations need to be met
and it is likely that the fast pyrolysis liquids will be classified as ''dangerous'' or
''hazardous'' substances in transportation. Pyrolysis liquids are not listed on the
UN approved carriage list [Anon 2000], and therefore its own classification has
been determined for them. The classification derived is:
UN 1993 Flammable Liquid, N.O.S., (Fast Pyrolysis Liquid), 3, 1º(a), 2º(a), 1
15
This code is valid for pyrolysis liquids containing less than 10 wt% acetic acid.
A separate classification is required for liquids with an acetic acid content higher
than this, although up to date no fast pyrolysis liquids with a content higher than
approximately 5wt% have been reported.
Label requirements for fast pyrolysis liquids
All packages of any size should be clearly marked with the information given in
Table 3, both on the sample and on the relevant transportation documents. A set
of Material Safety Data Sheets (MSDS) must also accompany any samples (Appendix A).
Table 3. Label information for packages.
UN symbol:
Substance
ID No.
Name of
substance
1993
Flammable
liquid
[Fast pyrolysis liquid]
Class 3 Flammable liquid
Label 3
Hazard
identification No.
33
Class 6 Toxic substances
Label 6.1
Label
model
Nos.
3
6.1
11
Class and item
numbers
3
1º(a), 2º(a),
2º(b), 3º(b),
5º(c)
Store this way up
Label 11
Figure 1. Packaging labels for fast pyrolysis liquids. All should be used on all
shipments nationally and internationally.
16
Table 4. Label sizes for shipments.
Capacity of package
Not exceeding 3 litres
Exceeding 3 litres but not exceeding 50 litres
Exceeding 50 litres but not 500 litres
Exceeding 500 litres
Exceeding 3000 litres
Dimensions of label
At least 52 x 74 mm
At least 74 000 x 105 000 mm
At least 105 x 148 mm
At least 148 x 210 mm
At least 250 x 250 mm
If the size of the package requires, the dimensions of the label may be reduced,
provided they remain clearly visible. IBCs [intermediate bulk containers] are not
suitable for pyrolysis liquids transport, due to the potentially hazardous nature of
the liquids. In addition, for tankers, or other large bulk transport, placards typical
of road and rail transports are used. The placard dimensions are typically at least
30 cm high and 40 cm wide, numerals to be at least 10 cm high. The requisite
codes for a placard are:
Substance Identification
No. [lower part]
1993
Name of substance
Flammable Liquid
[Fast Pyrolysis Liquid]
Hazard Identification
No. [upper part]
33X
A placard for large quantities [> 500 l] with the correct transportation codes is
shown in Figure 2.
33X
1993
Figure 2. Placard for transportation in containers and bulk carriage.
17
4. Homogeneity and sampling
4.1 Homogeneity of pyrolysis liquids
Pyrolysis liquids contain compounds with different chemical functionalities. The
homogeneity of the liquids is connected with the complex solubility and reactivity of these various chemical compounds in the liquid. Typically, pyrolysis liquids are single-phase liquids (Appendix B/3) containing varying amounts of
solids due to feedstock, process, and product recovery (see Chapter 2.1). The
heavy phase, containing entrained solids of the pyrolysis liquid, gradually settle
on the bottom of the container forming a sludge. The degree of sedimentation
depends on the density difference of the liquid and the particle size [Särkilahti
1996]. If the water content of pyrolysis liquid exceeds a certain limit (typically
30 wt%), the liquid separates into two phases (Appendix B/4). This type of
phase separation happens typically when the moisture content of the feedstock
has been too high (>15wt%).
Due to the inherent properties of some feedstocks, the liquid products may on
collection, exhibit two or more phases. Typical feedstocks that can give multiple
phases on pyrolysis, are forest residue, barks, pine [thin scum layer due to terpenes in biomass], some straws [e.g. wheat and rape straw], tropical hardwood
species [mahogany], eucalyptus and rape meal. These biomass types with either
high levels of hydrocarbon-soluble extractives or alkali metals are more prone to
give multiple phases. These phases cannot typically be mixed together and they
should be analysed separately.
Forest residue yields a pyrolysis liquid with a 10–20 wt% extractive-rich top
phase of low in polarity. The bottom phase, its amount being 80–90 wt%, is
similar to white wood liquids. The amount of top phase depends on the feedstock composition (extractives), process, and product collection conditions. The
top phase is, compared to bottom phase, low in water and density (see 6.1, Figure 4), and high in heating value (see 6.6, Figure 7) and solids content (Appendix D/6). In addition, it is slightly a non-Newtonian liquid (Figure 5). Some extractives (e.g. C18-C26 fatty acids) appear as crystals in the liquid.
18
4.2 Homogenisation and sampling
Mixing and sampling methods depend on the type and size of the pyrolysis liquid container (Appendix B). Attention should be paid to proper mixing of the
liquid. Small samples can be homogenised in routinely used laboratory mixers.
The pyrolysis liquid in barrels or in large totes can be homogenised by propeller
mixers. In case of a fresh single-phase low-to-medium viscous pyrolysis liquid
the liquid stays, under proposed storage conditions, homogenous for 6 to 12
months without any mixing. Only solids in the liquid settle gradually to the bottom. Circulation pumping or propeller mixing homogenizes the solids content in
the liquid [Oasmaa et al. 1997]. Round pumping can be employed if taken care
that the whole liquid is in movement. Heating of the liquid to > 50°C for longer
periods should be avoided to prevent evaporation of volatiles and/or viscosity
increase due to the chemical instability of the liquid.
In case of aged or poor-quality liquids, continuous stirring, emulsion stirring
(with a top-mounted slow revolving propeller stirrer) must be used during storing to avoid stratification/separation. Ordinary stirring methods, such as round
pumping (with heating), are not sufficient for inhomogeneous pyrolysis liquids
for creating storage circulation. Channelling, separation of light compounds, or
heavy (sediment) material concentrating in the round pumped stream can appear.
Handling, storage, and pumping may all affect liquids in different ways.
An aged phase-separated pyrolysis liquid can only be homogenised by adding
polar solvents, like alcohols (Appendix B/5). Acids, like acetic acid, also homogenise the liquid, but decrease the pH.
The homogeneity of the liquid after mixing can be verified by microscopic determination and/or by sampling from different depths and analysing the moisture
content (Appendix B). The sampling device should be wide-mouthed for obtaining a homogenous sample.
19
5. Composition
5.1 Water
Water is regulated in petroleum fuels, because it forms a separate phase, which
can cause corrosion, trouble in burners, or emulsion formation. Typically, water
is dissolved in pyrolysis liquids. The dissolved water in the fuel can increase the
efficiency of the boiler at levels of up to 15% and efficiency losses are not significant, until levels of 20% or higher are reached (Diebold et al. 1997).
The water content of pyrolysis liquids has been typically analysed by Karl
Fischer (KF) titration. Pyrolysis liquids contain low-boiling (below 100°C) water-soluble compounds, and hence conventional drying methods cannot be used.
The amount of volatile water-soluble compounds in the pyrolysis liquid is high,
and hence xylene distillation [ASTM D 95] in which the water is distilled away
with a co-solvent, cannot be used.
Earlier [Oasmaa et al. 1997] two standard methods, ASTM D 1744 and ASTM E
203, with several variables have been tested for hardwood, softwood, and straw
pyrolysis liquids. It was concluded that both KF titration methods can be used
for determination of water for pyrolysis liquid. The most important factor is
good dissolution of the sample. In 2000 the D 1744 has been removed from
ASTM standards without a replacement. The referred standard therefore is: Test
Method E 203-96: Standard Test Method to Water Using Volumetric KF Titration.
In order to determine the water content by Karl Fischer volumetric or coulometric titrations are used. The coulometric titration is developed for analysing trace
amounts of water. By proper choice of the sample size, KF reagent concentration
and apparatus, the volumetric titration is suitable for measurement of water over
a wide concentration range, that is, parts per million to pure water.
In volumetric KF titration the method is calibrated by determining the water
equivalent. It is usually quoted in mg of water per ml of KF solution:
Titre t = (mg of water)/(ml of KF solution)
20
(1)
In the titration the sample is dissolved in a suitable solvent (e.g. chloroform:methanol = 0..3:1) and titrated by a KF reagent. The reagent (e.g. 2methoxy ethanol) contains the reactive components: anion of the alkyl sulfurous
acid, iodine, and base (e.g. imadazol, pyridine). In the titration the anion of the
alkyl sulfurous acid reacts with the alcohol and forms ester. The ester is neutralized by the base (2):
ROH + SO2 + RN -> [RNH]SO3CH3
(2)
The anion of the alkyl sulfurous acid oxidizes to alkyl sulphate by iodine (3).
This reaction consumes water. For staying in optimum pH range (5-7) imadazol
is used.
H2O + I2 + [RNH]SO3CH3 + 2 RN -> [RNH]SO4CH3 + 2[RNH]I
(3)
ROH
An alcohol for instance, methanol, ethanol, ethylene-glycolmono-ethyl-ether
RN
A caustic solution, for instance imadazol or pyridine
As the use of chlorinated solvents is not recommended, methanol may be considered as a solvent when the dissolution of the sample is not a problem. Various
solvent combinations were tested for forest residue liquid (Appendix C/3): Chloroform:Methanol in ratios of 3:1, 1:1, 1:3, and pure Methanol. There was no
difference in water contents. However, the titration end-point was easier to detect when chloroform was present. Hence, the sample solvent for KF titration
[ASTM E 203] is recommended to be a mixture (3:1) of methanol and chloroform. It must be noted that also new KF reagents, including those containing no
pyridine, are on the market. In order to verify the use of another sample solvent
or reagent the use of water addition method (C/1) for calibration is recommended.
If the sample has dissolved properly, the fading titration end-point may be due to
a reaction of ketones and aldehydes [Riedel-de Haën 1995] with the titrant. Aldehydes and ketones may form acetals and ketals for example with methanol
yielding water as reaction product. This water is then titrated yielding to too high
water content. Aldehydes are more prone to this reaction than ketones. One solu-
21
tion for this is to use HYDRANAL K reagents [Riedel-de Haën 1995], which
prevent this reaction. Other error sources include: impropriate homogenisation
and/or sampling, too small sample size both for the sample and for the water
equivalent, wet drying agent, wet titration solvent, air-leakage, dirt on electrode,
dirty solvent, and sample deposits on the walls of the sample vessel (Appendix
C/2).
The accuracy of titration for good quality white wood liquids is ±0.1 wt% [Oasmaa et al. 1997]. For extractive-rich forest residue liquids 0.5 wt% deviation of
duplicates should be accepted.
Attention should be paid to proper sample homogenisation. A sample size of
0.25 g is recommended for pyrolysis liquids containing about 20 wt% water. The
final water content is calculated based on the water equivalent of the titrant and
the consumption of the titration reagent. For a good quality pyrolysis liquid the
0.01 mg/ml deviation in water equivalent can be accepted. A variation of 5.70–
5.73 mg water/ml titrant in the water equivalent yields a variation of 28.50–
28.65 wt% in the water content (0.25 g sample, 12.50 ml consumption of titration reagent).
5.2 Solids
There are varying amounts of solids in pyrolysis liquids due to feedstock, process, and product recovery (see chapter 2.1). The solids content of pyrolysis liquids affected particulate emissions observed in combustion tests in a commercial
heavy fuel oil boiler at Birka Energi (former Stockholm's Energi), Sweden,
[Hallgren 1996], in a commercial heavy fuel oil/light fuel oil boiler at Oilon,
Finland [Oasmaa & Kytö 2000], and at Fortum [Gust 1997] and Sandia National
Laboratories [Suppes et al. 1996].
Various solvents (ethanol, methanol, acetone) for solids determination have been
tested previously [Oasmaa et al. 1997]. The main observations were: 1) Ethanol
was a better solvent than acetone for pyrolysis liquids from hardwood
(oak/maple, eucalyptus). 2) Filtration of the acetone solution was more difficult
(long filtration time, sticky filter cake) than that of the ethanol solution. 3) For
the straw pyrolysis liquid, a clear difference (10 wt%) in solids contents was
22
observed when determined using ethanol and acetone. Acetone either did not
dissolve the straw pyrolysis liquid properly or reacted causing precipitation. 4)
The filter pore size (0.1–10 µm) did not affect the amount of solids retained except for the case of straw liquid, which contained high amounts of submicron
particles. 5) The differences in solubility in ethanol and methanol were very
small. Methanol was a slightly more effective solvent than ethanol for pyrolysis
liquids from pine, eucalyptus, and straw. For an oak/maple liquid, no difference
in methanol and ethanol insolubles was obtained.
A number of effective polar solvents, like methanol and ethanol, may be used for
measuring the solids content of pyrolysis liquids. At VTT, ethanol was chosen
for white wood liquids because its solubility power is high for several different
pyrolysis liquids, and it is safer for use in routine measurements than methanol.
However, in case of extractive-rich pyrolysis liquids like forest residue liquid, a
neutral solvent is needed with a polar one. For forest residue liquids (see 4.1), a
mixture of methanol and dichloromethane was effective (Appendix D). For bottom phase (low in extractives) of forest residue liquid 20 vol% dichloromethane
in methanol was the most effective solvent of those tested (D/4). For top phase
(high in extractives), the best solvents were (an order of preference D/6): Methoxy Propanol-Dioxane (1:1), Industol (PE2, a trade name, 88.3% ethanol, 2.7%
methyl-isobutylketone, 1.8% acetone)-Dioxane (1:1), Isopropanol (IPA)Dioxane (1:1), IPA-Dichloromethane (CH2Cl2) (1:1), Methanol- CH2Cl2 (1:1).
A 1 µm filter is used because of fine particles in the liquid. If the filtration time
is long (high solids content), a 3 µm filter can be used and the filtrate filtered
through a 1 µm filter. A sample size yielding 10–20 mg residue and a sample to
solvent ratio of 1 : 10 with several washings is recommended (Appendix D).
Maximum 10 wt% difference between duplicates can be accepted.
5.3 Particle size distribution
Two optical methods have been previously tested for determining particle size
distribution for pyrolysis liquids: a particle counter and an image analyser [Oasmaa et al. 1997]. Observations as follows were made: 1) A dark colour of the
pyrolysis liquid may disturb the detection. 2) Overlapping of several particles
23
can be detected as one large particle. 3) The flow rate may have some effect on
the particle size distribution.
A method for measuring particle size distribution of pyrolysis liquid is to use
microscopy with the particle counter model. At VTT, Leica DM LS microscopy
with Leica Qwin Lite Standard photo analysis programme is used. In the system,
the particles are marked and the computer programme measures, calculates and
classifies them. In Figure 3 (see also Appendix E), the particle size distribution
(1 µm and above) of forest residue liquid (whole liquid, solids ~ 0.5 wt%) based
on maximum length of particle is presented. The method gives particles below 1
µm as "undersize particles". The numerical amount of these submicron particles
is large, but the volumetric amount small (Appendix E). The reproducibility
(triplicates) of the determination is good. If the solids content is high, overlapping of particles may cause error in the results.
35
Share of total %
30
25
1
2
3
20
15
10
5
0
1.58 2.51 3.98 6.31 10 15.8 25.1 39.8 63.1 100
Length of the particle µm
Figure 3. Particle size distribution of forest residue liquid. Solids 0.47 wt% as
methanol-dichloromethane insolubles.
24
5.4 Elemental composition
Elemental analysis of carbon, hydrogen, and nitrogen was carried out at VTT
with LECO CHN600 according to ASTM D 5291. In the method, carbon, hydrogen, and nitrogen are simultaneously determined as gaseous products (carbon
dioxide, water vapour, and nitrogen).
The accuracy of carbon and hydrogen is good, but poor for nitrogen [Oasmaa et
al. 1997]. This is due to low concentrations of nitrogen (≤0.1 wt%) in wood liquids and to the low nitrogen detection limit (0.1 wt%) of the method. Pyrolysis
liquids from straw and forest residue contain higher (0.2–0.4 wt%) concentrations of nitrogen and hence stdevp% is lower [Oasmaa et al. 1997]. Because of
the small sample size, the reproducibility of the elemental analysis is dependent
on the homogeneity of pyrolysis liquids. At least triplicates are recommended, if
the sample is inhomogeneous.
The sulphur content of wood pyrolysis liquids is typically very low (60–500
ppm). The straw liquid has also a low sulphur content (400–500 ppm). The detection limit for sulphur analysis by LECO SC 32 according to ASTM D 4208 is
0.1 wt%. Lower concentrations can be analysed, e.g., by ICP (Inductively Coupled Plasma). The detection limit for analysing sulphur by ICP is about 5 ppm
due to wet oxidation as the pretreatment method. Low sulphur contents can be
analysed by capillary electrophoresis technique after combusting the sample in
an oxygen bomb according to ASTM D 4239. Simultaneously, the total chlorine
is also given. The chlorine content of pyrolysis liquids from different feedstocks
has been below 5 ppm for hardwood (oak maple) liquid, 75 ppm for softwood
(pine) liquid, and 300–400 ppm for forest residue (high content of green material) and straw (wheat) liquid. Oxygen (dry basis) is calculated by difference.
5.5 Ash and metals
Ash or char is present in the pyrolysis liquids due to inefficiencies in char removal equipment such as cyclones leading to carryover of char into the liquid
collection system. It is the inorganic part of solids presented earlier containing
unburnt carbon, and metals. It may also contain sand from process heat carrier.
Depending on the use of the fuel, ash content and composition has a consider-
25
able bearing on whether or not detrimental effects will occur. High ash content
can cause high wear in pumps and injectors, deposits in combustion equipment,
or corrosion in gas turbines due to alkali metals in ash. Large boilers and lowspeed engines can tolerate more ash than more sophisticated equipment.
Ash content is measured according to DIN EN 7. In the standard method, the
sample is ignited and burnt in a crucible (Pt, quartz or porcelain). Carbon containing residue is ashed at 775°C, cooled, and weighed.
A high amount of water in pyrolysis liquids can cause foaming or splashing during heating of the sample. Hence, a controlled evaporation of water is needed. A
sample of 20 ml pyrolysis liquid may be let to heat in a porcelain crucible on a
heating plate or in a temperature-controlled small sand bath for evaporation of
water and other volatile components. Addition of isopropanol or ash-free filter
paper for absorbing the water can prevent splashing.
Part of ash consists of alkali metal oxides. As the alkali content of wood pyrolysis liquids is low, possible evaporation during ashing may not be important for
the total ash. In case of pyrolysis liquids from straw and grasses the situation
may be different. These liquids contain much higher amounts of alkali metals.
During ashing, part of alkali metals may evaporate and the rest may form oxides
that can lead both to underestimation (evaporation) and overestimation (oxidation) of the ash content.
Ash contents have typically been low (0.1–0.2 wt%) for white wood pyrolysis
liquids and higher (0.2–0.4 wt%) for forest residue and straw liquids. For hotvapour filtered pyrolysis liquids (hardwood, softwood, straw), low ash contents
(0.01 wt%) have been obtained due to efficient removal of char fines with alkali
(Na, K, Ca, Mg) metals (below 10 ppm) [Oasmaa et al. 1997].
For alkali metal analyses, different sample pretreatment methods for ICP-AES
(Inductively Coupled Plasma Atomic Emission Spectrometry) and AAS (Atomic
Absorption Spectrometer) have been tested previously [Oasmaa et al. 1997]. The
straight dissolution in isopropanol (IPA) was inaccurate due to incomplete dissolution of the sample. Dry combustion was time-consuming, because the water
has to be evaporated carefully before ashing, or else the liquid would swell out
from the Pt crucibles. If a hard crust forms on the surface of the sample before
26
the end of drying, the water may suddenly vaporise and the sample is lost. Wet
oxidation was the easiest and fastest method. However, large samples are difficult to handle with acids, and a sample size of 3–5 g is suggested as adequate.
The same acid or acid mixture and concentrations should be used both in the
sample and in the standard solutions.
Similar concentrations of alkali metals (Na, K, Ca, Mg) were obtained with both
dry combustion (ashing temperature 520°C for 2 hours) and wet oxidation as
pretreatment method. Stdevp% was reasonably low (0–14) when using sample
sizes of 3–5 g even for alkali concentrations of 10 ppm.
Metals may be analysed by ICP-AES or AAS. If all metals are to be analysed,
ICP-AES is suggested as an easier (several metals from one run) and hence,
cheaper method. However, AAS is more sensitive in some cases. Wet oxidation
(Appendix F) is suggested as a fast and easy pretreatment method especially for
analysing alkali metals and easily volatilised toxic metals like Cr, As, Pb, Hg,
and Cd.
It should be pointed out that for accurate analysis of trace alkali metals the
whole procedure from the recovery of the pyrolysis liquid to the sample pretreatment should be re-checked in detail. Contaminations from glass containers
and from dust in air may have an influence on the results [Diebold et al. 1997].
The use of Teflon (polytetrafluoroethylene, PTFE) bombs for sample pretreatment should be considered. If traces of alkalis are to be analysed, the use of a
method requiring no sample pretreatment, like neutron activation (NA) analysis,
may be advantageous.
27
6. Fuel oil properties
6.1 Density
Specific gravity is of little significance as an indication of burning characteristics, but is used in calculating weight/volume relationships, e.g., the heating
value [Diebold et al. 1997].
The density is measured according to ASTM D 4052 at 15°C by a digital density
meter. A small volume (approximately 0.7 ml) of liquid sample is introduced
into an oscillating sample tube and the change in oscillating frequency caused by
the change in the mass of the tube is used in conjunction with calibration data to
determine the density of the sample. An Anton Paar DMA 55 density meter was
used at VTT. A viscous pyrolysis liquid with large particles may disturb the
measurement and cause erroneous results. In addition, air bubbles may disturb
the determination near ambient temperatures. Therefore, vigorous shaking of the
sample just before the analysis should be avoided. Instead, the sample can be
rotated carefully. At elevated temperature (50°C) the air bubbles are easily
avoided by preheating the sample for a short period in a closed vessel.
The density of pyrolysis liquids is typically 1.2–1.3 kg/dm3. The precision of
density measurements is good (variation below ±0.1%). Densities of wood pyrolysis liquids are a function of water content [Oasmaa et al. 1997]. In Figure 4
densities of various pine and forest residue liquids (top : bottom = 1 - 2 : 8 - 9)
are shown as a function of original water content. It can be seen that the bottom
phase of forest residue liquid has similar density-water correlation than pine
liquids. The density of the top phase diverges from that of the bottom phase,
when the water content of the top phase decreases. This is due to the concentration of extractives in the top phase, while water with water-soluble compounds
moves to the more polar bottom phase.
28
Density @15C, kg/dm3
1,26
Pine
1,24
1,22
1,20
1,18
FR,Bottom
1,16
1,14
FR,top
1,12
1,10
10
15
20
25
Water, wt-%
30
35
Figure 4. Density of pine and forest residue (FR, top : bottom = 1 - 2 : 8 - 9)
pyrolysis liquids as a function of water content.
6.2 Viscosity
Viscosity is a measure of the resistance of the liquid to flow. The viscosity of the
fuel is important, i.e., because of its effect on pumping and injecting of fuel [Dyroff 1993].
The viscosity of pyrolysis liquids can be determined as kinematic viscosity using
glass capillaries or as dynamic viscosity using rotational viscometers. The correlation between the kinematic and dynamic viscosity can be presented by the
following equation:
η [cSt] = η [mPa •s] /ρ [kg/dm3],
where
ν
η
ρ
kinematic viscosity at temperature T
dynamic viscosity at temperature T
density of the liquid at temperature T.
29
(4)
The viscosity of standard fuels, which are Newtonian liquids, is typically measured as kinematic viscosity according to ASTM D 445. In the standard method,
the time is measured in seconds for a fixed volume of liquid to flow under gravity through a calibrated capillary at a closely controlled temperature. The kinematic viscosity is the product of the measured flow time and the calibration constant of the viscometer.
Leroy et al. [1988] have carried out an extensive study on rheological characterisation of several different pyrolysis liquids. They concluded that all pyrolysis
liquids exhibited an essentially Newtonian behaviour in the range of shear rate
examined (101 to 103 s-1). For Newtonian liquids, the viscosity stays constant
when increasing the shear rate. Additional tests were carried out by VTT at Haake Application Laboratory in Karlsruhe, Germany, using an RT 20 all-round
rheometer with mechanical bearing. Haake concluded that the samples were
Newtonian liquids.
Tall oil soap, which mainly consists of extractives, is a non-Newtonian material.
Due to the high extractive content of forest residue liquids, its Newtonian behaviour was tested (Figure 5). The bottom phase is chemically similar to the white
wood pine liquid and its Newtonian behaviour was not a surprise. The top phase
containing most of the extractives showed a slight non-Newtonian behaviour at
30°C. At 45°C, the non-Newtonian behaviour was undetectable.
Kinematic viscosity can be determined by using Cannon-Fenske or Ubbelohde
capillaries. The basic difference between these methods is the flow direction of
the sample. For non-transparent liquids, Cannon-Fenske is easier to use near
room temperature because of up-flow system and wider capillaries. In the Ubbelohde method, the sample flows downwards and the capillary bore is much
smaller. This may lead to an uneven flow of viscous pyrolysis liquid in the capillary.
30
Viscosity, mPas
1600
1400
1200
1000
800
600
400
200
0
0
30 °C bottom
45 °C bottom
30 °C top
45 °C top
25
50
75
100
125
Shear rate, 1/s
Figure 5. Dynamic viscosity (η) as a function of shear rate (τ) for the two
phases (top and bottom) of forest residue liquid (water content: top 6.8 wt%,
bottom 24.1 wt%).
The comparison of the two capillaries was carried out for one hardwood pyrolysis liquid. At 20°C, there was a small variation between viscosity results obtained either by Cannon-Fenske or Ubbelohde method. At 50–80°C, no significant difference was observed in the viscosity of pyrolysis liquid measured by
these methods. The small difference in viscosities measured by the two capillary
methods according to ASTM D 445 may be explained by difficulties in reading
the volume in the Ubbelohde method (downward flow) [Oasmaa et al. 1997].
Another reason for this may be that a small error in temperature causes higher
error at 20°C than at 40°C (Figure 6.)
31
150
140
130
120
110
Viscosity, cSt
100
90
80
70
60
50
40
30
20
10
1 °C
0
0
10 20 30 40 50 60
Temperature, °C
Figure 6. Viscosity curve of a softwood pyrolysis liquid.
The determination at near-ambient temperature may be disturbed by air bubbles,
especially for forest residue liquids, and therefore vigorous shaking of the sample just before the analysis should be avoided. The sample can be instead rotated
carefully. At elevated temperature the air bubbles can easily be removed during
preheating of the liquid.
When comparing the kinematic viscosities with the dynamic viscosity obtained
by a rotaviscotester (Haake VT 550 controlled rate rotaviscotester, 8 ml NV cup,
32
46 ml MV-DIN cup, Tmax 100°C) the precision was good below 50°C. At 80°C,
the evaporation of volatiles was observable already during sample equilibration.
The viscosity of the sample increased by almost 1% in a minute during measuring at 80°C. The stdevp at 80°C was also higher than by using the capillary system, where the evaporation of volatiles is not as significant due to the small open
surface of the capillary. The equilibration times of 15 minutes for 20°C and
40°C, and 10 minutes for 80°C were used [Oasmaa et al. 1997].
Viscosity, cSt
Viscosities of pyrolysis liquids are a function of water content [Oasmaa et al.
1997]. Viscosities for various pine and forest residue liquids are shown in Figure 7.
40
35
30
25
20
15
10
5
Pine 40°C
Pine 50°C
FR, bottom
50°C
Linear (Pine
50°C)
15
20
25
30
Water, wt-%
35
Figure 7. Viscosity of pyrolysis liquids from pine and forest residue (FR, bottom
phase).
6.3 Pour point
The pour point of a fuel is an indication of the lowest temperature, at which the
fuel can be pumped [Dyroff 1993]. The upper limit for pumpability is about
600cSt [Rick & Vix 1991]. The pour point was determined according to ASTM
D 97. In the method, a preheated sample was cooled at a specific rate and examined at intervals of 3°C for flow characteristics. The lowest temperature, at
which oil movement was observed, was recorded as the pour point. The setting
point is the temperature, at which the oil cannot be pumped, and it is 2–4 K
lower than the pour point.
33
Preheating of the sample was laborious for some pyrolysis liquids, as it may
cause separation of water on the surface of the sample. The water froze before
the actual pour point and hence disturbed the measurement. The pour point for
wood (hardwood, softwood, forest residue) pyrolysis liquids ranged from –12°C
to -33°C. For straw (wheat) liquids -36°C was measured for two different
batches. For the pyrolysis liquids analysed the low viscosity was an indication of
a low pour point.
6.4 Cloud point
Under low temperature conditions, paraffin constituents of a mineral fuel oil
may precipitate as a wax, forming a cloud in the liquid. The wax settles out and
blocks the fuel system lines and filters causing malfunctioning or stalling of the
engine. The temperature, at which the precipitation occurs, depends on the origin, type, and boiling range of the fuel. The more paraffin in the fuel, the higher
the precipitation temperature and the less suitable the fuel for low temperature
operation [Dyroff 1993].
The cloud point is determined as the temperature, at which a cloud of wax crystals first appears in a liquid, when it is cooled under conditions described in the
method ASTM D 2500. Despite low levels of paraffin compounds in pyrolysis
liquids, the cloud point was measured for the pyrolysis liquid according to this
method to check, if other components can cause a similar phenomenon. The
sample was cooled at a constant rate down to -21°C, at which the sample was no
more fluid. No "clouding" was observed, which may be due to the dark colour of
pyrolysis liquid. It was concluded that this test is not suitable for dark pyrolysis
liquids [Oasmaa et al. 1997].
6.5 Conradson carbon residue
The carbon residue is a measure of carbonaceous material left in a fuel after all
the volatile components are vaporised in the absence of air. There is no real correlation between Conradson carbon results for diesel fuels and deposit formation
on injector nozzles. Fuels with up to 12 wt% Conradson carbon residues (CCR)
have been used successfully in slow speed engines [Dyroff 1993]. The signifi-
34
cance of the Conradson carbon test results also depends on the type of the engine, in which the fuel is used. Pressure jet and steam atomising type burners are
not very sensitive to the carbon residue of the fuel used.
The Conradson carbon residue was measured according to ASTM D 189. In the
method, a weighed quantity of sample is placed in a crucible and subjected to
destructive distillation. The residue undergoes thermal cracking and coking reactions during a fixed period of severe heating. At the end of the specified heating
period, the crucible containing the carbonaceous residue is cooled in a desiccator
and weighed. The residue is calculated as a percentage of the original sample,
and reported as CCR.
Due to the high water content of pyrolysis liquids, causing foaming, the standard
sample size was reduced from 3–5 g to 1 g. For heterogeneous liquids, a larger
sample size (3–5 g) is needed, and the water should be evaporated carefully from
the crucible on a heating plate to avoid spilling of the liquid sample.
The accuracy of the method according to the standard is ± 2 wt%, when CCR is
above 20 wt%. The CCR has typically ranged 18–23wt% (22–23 wt% on dry
pyrolysis liquid) for pyrolysis liquids from hardwood and softwood and 17–18
wt% (22–23 wt% on dry basis) for pyrolysis liquids from forest residue or wheat
straw. Low CCR values (17–22 wt% on dry basis) have been determined for hotvapour filtered pyrolysis liquids from hardwood (poplar), softwood (pine), and
straw (wheat) [Oasmaa et al. 1997]. This may be an indication of change in the
chemical composition of the liquid, e.g., sticking and possibly also thermal
cracking of heavy compounds on the filter cake during hot vapour filtration, as
the amount of solids alone does not explain it.
Normally Conradson Carbon is only specified for light diesel fuels. The pyrolysis liquids are of different chemical nature compared to mineral oil based heavy
diesel fuels. There is not yet adequate information of diesel performance data for
pyrolysis liquids, and therefore also this test was selected for evaluation [Oasmaa et al. 1997]. CCR can only be taken as a qualitative comparison between
pyrolysis liquids.
35
6.6 Heating value
The heat of combustion of the fuel is the amount of heat produced, when the fuel
is burned completely. It may be determined by bomb calorimetric techniques.
There are two values for the heat of combustion, or calorific value, for every
fuel. They are referred as the gross (or HHV, higher heating value) and net (or
LHV, lower heating value) heats of combustion. The difference between the two
calorific values is equal to the heat of vaporisation of water formed by combustion of the fuel.
The heating value is measured as HHV by DIN 51900. A Parr adiabatic bomb
with a Parr Calorimeter Controller 1720 is employed. The heat of combustion is
determined by measuring the temperature increase in the water jacket and then
calculated from the energy balance for the system. The high water content of the
pyrolysis liquids may lead to poor ignition, and therefore a fine cotton thread can
be used as a wick. The heat content of the thread is subtracted from the result.
The lower heating value (LHV) is calculated from HHV and the hydrogen content [ASTM 529192] by equation (5). No subtraction of free water has to be
done [Rick & Vix 1991] because the water in the pyrolysis liquid cannot be removed by centrifugation as for heavy petroleum fuel oils.
LHV [J/g] = HHV [J/g] - 218.13 x H% [wt%]
(5)
The heating values correlate with the water content of pyrolysis liquids [Oasmaa
et al. 1997]. The LHVs of pine and forest residue liquids are shown in Figure 8.
The heating value of the bottom phase of the forest residue liquid correlates with
the water similar to that of pine liquids. The maximum water content for forest
residue liquids (bottom phase) is around 33 wt-% after which a phase separation
happens. The heating value of the extractive-rich top phase is higher than that of
the bottom phase. The difference is larger for lower water contents (higher extractive content in the top phase). The heating value of the pyrolysis liquid is
roughly half (of dry matter) that of petroleum fuels (Table 1).
36
LHV, MJ/kg
30
28
26
24
Pine
22
20
18
FR,
bottom
16
FR, top
14
12
10
5
10
15
20
25
Water, wt-%
30
35
Figure 8. Heating value of pine and forest residue liquids (FR) as a function of
water content.
6.7 Flash point
The flash point of petroleum oil is measured to indicate the maximum temperature, at which it can be stored and handled without serious fire hazard. Too low a
flash point causes the fuel to be subject to flashing, and possible continued ignition and explosion. In spite of its importance from a safety standpoint, the flash
point of the fuel is of no significance to its performance in an engine. The
autoignition temperature is not generally affected by variations in flash point,
neither are other properties, such as fuel injection and combustion performance
[Dyroff 1993].
The flash point can be determined according to ASTM D 93 using a PenskyMartens closed-cup tester [ASTM D 93/IP 34]. The sample is heated at a slow,
constant rate with continual stirring. A small flame is directed into the cup at
regular intervals with simultaneous interruption of stirring. The flash point is the
lowest temperature, at which the test flame ignites the vapour above the sample.
Flash points from 40°C up to above 100°C have been measured for pyrolysis
liquids [Oasmaa et al. 1997]. The low flash point of pyrolysis liquids (40–50°C
for forest residue liquids) is due to a high amount of low-boiling volatile com37
pounds. A low flash point also indicates a high vapour pressure. For liquids low in
volatile, the flash point is much higher (above 100°C). The flash point cannot be
measured at 70–100°C, where the evaporation of water suppresses the ignition.
6.8 Ignition temperature
Unlike the petroleum-based fuels, the pyrolysis liquids do not spontaneously
ignite in a typical compression ignition engine. The ignition temperature for
aromatic-rich oils is much higher than for paraffin-based fuels [Rick & Vix
1991]. In addition, the high water and oxygen contents with a substantial amount
of non-volatiles contribute to poor ignition.
The cetane number is used for light oils and diesel fuels and can be measured
according to ASTM D 613. In the method, the tested fuel is ignited by compression, the ignition delay is measured and the results are then compared with those
of known fuels. Autoignition of a pyrolysis liquid has been tested in a single
cylinder diesel engine instrumented for cylinder pressure analysis [Solantausta et
al. 1993]. The pyrolysis liquid ignited only after adding ignition-improver.
Properties of pyrolysis liquids vary significantly depending on feedstock, process, and product recovery. Hence, autoignition with a higher-quality pyrolysis
liquid may succeed as stated by Shihadeh [Ph.D.Thesis]. Shihadeh’s work was
carried out with a combustion bomb to simulate the diesel engine environment,
but not with a diesel engine.
The Ignition Quality Tester (IQTTM) is a combustion-based analytical instrument, by which the ignition delay and cetane number (CN) of diesel and alternative fuels (including cetane improved diesel fuels) can be accurately determined.
The American Society for Testing and Materials (ASTM) has approved the establishment of a study group to implement an ASTM standard test method for
use of IQTTM to determine the cetane number of diesel fuels. CanMet and AET
(Advanced Engine Technology Ltd.) in Canada have carried out IQT tests with
pyrolysis liquid emulsions. Based on these results and by linear interpolation
they ended up to a cetane number of 5.6 for the tested hardwood pyrolysis liquid
[Ikura et al. JOR3-CT98-0253, Final report].
38
Instituti Motori (Italy) has studied the combustion characteristics of pyrolysis
liquids by evaluating the ignition delay as the time lag between the start of the
needle lift and the start of combustion. By means of a Reference Fuel Correlation, the Cetane Number of about -10 was evaluated from the Ignition Delay
[Bertoli et al. JOR3-CT98-0253, Final report].
The method for determination the ignition temperature of pyrolysis liquids is
under testing at VTT employing a thermobalance (TG).
The test developed to characterise gasoline properties (octane number) in spark
ignition engines is not appropriate for pyrolysis liquids, as significant improvements on the pyrolysis liquid quality would be required. Antiknock rating or the
octane number can be measured in a single-cylinder engine test according to
ASTM D 2699 - 68 (IP 237/69). The fuel used should possess the following
characteristics: high volatility, good stability in preheating (135 ± 8.5°C), good
miscibility with hydrocarbons, stability with air, neutrality, stability during the
test, a low amount of carbon deposit, and easy to clean. Pyrolysis liquid does not
fulfill these requirements.
6.9 Thermal conductivity, specific heat capacity and
electrical conductivity
Thermal conductivity and specific heat capacity are essential in the design and
evaluation of transport units and sizing process equipment, i.e., heat exchangers,
atomisers and combustors. There are two methods for measuring thermal conductivity [Jamieson 1975] – absolute or comparative. In the absolute method the
heat conducted across a film of the test fluid located in the annular space between two vertical copper cylinders is measured. Thermal conductivity of pyrolysis liquids was determined using the more common comparative method where
the heat conducted across a thin film of the test fluid in the space between a
nickel coated sphere and a surrounding block using a relative method is measured [as described in Peacocke et al. 1994].
The chemical reactivity of pyrolysis liquids leads to erroneous results and values
can vary even within batches of liquids, due to poor mixing and phase separation. Results from the work of Peacocke give an average thermal conductivity of
39
0.386 W/mK over the temperature range 44–63°C for mixed hardwood-derived
fast pyrolysis liquids.
Measurements for specific heat capacity of pyrolysis liquids were carried out
using a test-rig [Peacocke et al. 1994] where pyrolysis liquid was pumped
around a closed loop at approximately 0.1 g/s. The liquid passed through the cell
where it was heated and returned to the reservoir. Heat losses were minimised by
sealing the cell body under high vacuum and covering it with aluminium. The
pyrolysis liquid temperature change across the heater was measured. Power input was calculated by measuring the potential difference across the heater and a
thermally stable resistor connected in series with the heater. The mass flow rate
was measured at 2°C intervals by sampling the oil flow rate for two minutes.
The system was calibrated using Shell Thermia B oil. Results of the work of
Peacocke give an average value of 3.2 kJ/kgK [+300 J/kgK] over the temperature range of 26–61°C.
Electrical conductivity is a property that is of no direct use for fuel applications,
but is required by some instruments for level measurement and control. There
are no published data for values. However, Wellman Process Engineering Ltd.
has provided some data as indicated in Table 5.
Table 5. Electrical conductivity of pyrolysis liquids.
Sample code
DYN1002
BTG2G
BK40/90W7
Water content
wt%
28.54
22.97
21.33
Char content
wt%
1.49
0.77
--
Conductivity
µS/cm
50
60
200
Data supplied by Wellman Process Engineering Ltd. [measured using a standard electrical
conductivity meter] Water and char content data supplied by Aston University.
6.10 Lubricity
Viscosity does not describe the lubricating properties of the oils. Lubrication
properties are crucial, for example, for selection of supply pumps [Rick & Vix
1991]. Two methods measuring lubricating properties of diesel fuels have been
40
tested for pyrolysis liquids [Oasmaa et al. 1997]: Cameron Plint TE77 High Frequency Friction equipment, and a fourball wear test according to IP 239/69T
[ASTM D 278388] were used. The former method is not suitable, because it is a
totally open system and the evaporation of pyrolysis liquid disturbs significantly
the measurement. The latter method may be used. Considering the preliminary
tests, it seems that the pyrolysis liquids possess some lubricating properties. It
should be borne in mind that the viscosity of the sample affects the results and
no exact conclusions can be drawn due to the great difference in viscosity between pyrolysis liquids and diesel fuels [Oasmaa et al. 1997].
41
7. Unique properties of pyrolysis liquids
7.1 Acidity and material corrosion
The pH of pyrolysis liquids is low (2–3) due to high amounts (8–10 wt%) of
volatile acids, mainly acetic and formic acid [Sipilä et al. 1998, Fagernäs 1995].
These acids with water are the main reasons for the corrosiveness of pyrolysis
liquids especially at elevated temperatures [Aubin & Roy 1980].
When producing pyrolysis liquids an even black coating can be formed on the
steel surface of the process equipment. This “coating”, analysed by ESCA (Electron Spectroscopy for Chemical Analysis), contained metals from pyrolysis liquid, mainly silica, potassium, and calcium. No corrosion was observed.
The corrosion test specified for fuel oils is empirically based (colour changes for
copper strips) and bears little significance to the corrosion potential of a pyrolysis liquid. The colour changes observed, resulting from copper reacting with
different chemicals (primarily sulphur-containing compounds) found within the
petroleum. The calibration of this type of test or the development of an entirely
new test is required, before the corrosion potential of pyrolysis liquids can be set
down in a specification [Diebold et al. 1997].
In the standard corrosion test at 60°C (ASTM D 665 A), no rust was formed, but
a clear weight loss in carbon steel (AISI 01) was observed. Aubin and Roy
[1980] reported no corrosion for carbon steel at ambient temperature in low acid
(3.6%) and water (4.4%) concentrations, but they observed clear corrosion at
elevated temperature (45°C) in high acid (17.5%) and high water (55.7%) contents.
The acid-resistant steel AISI 316 (17% Cr, 11% Ni, 2.2% Mo, 0.05% C) can be
used. AISI 316 may be a little better than AISI 304 because of the small amount
of Mo that serves to make the steel more resistant to general corrosion in nonoxidising acids, to stress corrosion and especially to localised corrosion (pitting
and crevice corrosion) caused by aggressive components like halogens [Oasmaa
et al. 1997]. It is difficult to find suggestions for corrosion allowances for pyrolysis liquids in the literature. Rates have been suggested for ''pyroligneous acid –
wood distillation products'': < 0.0001 cm/y for 304 and 316 SS [Polar 1961].
42
Extensive testing of various metals (brass, mild steel, aluminium and stainless
steel) has been carried out by Orenda [Fuleki 1999]. Brass and stainless steel
were relatively unaffected, but aluminium and mild steel suffered severe weight
loss at higher temperatures and prolonged exposure (17.8 and 15.8 wt% loss
respectively during 360 hours, 70°C). If the surface deposits on the mild steel
were removed, then even higher corrosion rates were reported (19.6 wt% loss
during 360 hours, 70°C). It was noted that brass might not be suitable in fuel
combustion systems, due to the potential erosion by particulate matter in the
liquids. Soltes and Lin [1984] also reported that the pyrolysis liquids are corrosive to mild steel and aluminium.
Jay et al. [1995] carried out material testing related to Wärtsilä's diesel engine
tests. The nozzle would be a critical component with respect to abrasive wear
due to high particle content and corrosion. The tests in a 400 bar injection test
rig showed M390 to be suitable material for the nozzle. M390 is Martensitic
Sintered Stainless steel with a composition of 1.90% C : 20% Cr : 1% Mo : 4%
V : 0.6% W, which can be through hardened to achieve a 62 HRc and can withstand soak temperatures of up to 500°C. X90CrMoV18 (AISI 440B) stoff
1.4112 Martensitic Stainless steel, 57 HRc hardness is suitable for engine pushrods and injection needles [Jay et al. 1995].
Nickel and nickel-based materials are not resistant to pyrolysis liquids even at
room temperature. On the other hand, cobalt-based HAYNES 188 (39.4% Co,
22% Cr, 22.9% Ni, 14.5% W, 1.2% Fe) was resistant in the corrosion tests at
< 80°C [Oasmaa et al. 1997].
In the copper corrosion test (ASTM D 130) no corrosion or weight loss on copper stem (99.9% electrolytic copper) was observed for different pyrolysis liquids
(hardwood, softwood, straw) at 40°C. Copper is a noble metal and hence has
generally a good corrosion resistance to non-oxidising acids. Copper is suitable
for washers [Jay et al. 1995]. However, if connected with other metals, there is a
possibility of electrochemical corrosion. Copper and its alloys (brass, bronze,
cupronickel) are widely used in piping applications (tubes, valves, etc.) mostly
because of the excellent availability of different components. However, it should
be borne in mind that copper and brasses are subject to erosion and corrosion,
when high fluid velocities are used or abrasive particles are present especially at
43
higher temperatures. It should also be pointed out that brass with 15% Zn or
more could not be used with pyrolysis liquids due to dezincification.
Many plastics like PTFE (polytetrafluoroethylene), PP (polypropylene), PE
(polyethylene), HDPE (high density polyethylene), and polyester resins are very
resistant to pyrolysis liquids [Oasmaa et al. 1997, Czernik 1994]. They are excellent materials for containers in storing, transportation and sampling of pyrolysis
liquids. EPDM (Ethylene Propylene Diene Monomer rubber) and Teflon O rings
can be used in sealing [Jay et al. 1995]. Viton O rings react with pyrolysis liquid
causing material expansion [Jay et al. 1995]. The use of plastics could possibly
be extended to replacing copper or, in special cases, also AISI 316.
7.2 Solubility
7.2.1 Water solubility
Water is chemically dissolved in pyrolysis liquids [Oasmaa et al. 1997]. Pyrolysis liquids can be considered as mixtures of water and water-soluble organic
compounds with water-insoluble, mostly oligomeric material (Table 6). The
ratio of these fractions depends on the feedstock, on the process and storage
conditions. The lignin-derived oligomers usually account for 30–40 wt% of the
liquid, while water concentration ranges from 20 to 30 wt%. In such cases, pyrolysis liquids are typically in single phase because of the presence of polar carboxyl and hydroxyl compounds. However, phase separation can take place in
higher water and/or lignin-derived material concentrations [Oasmaa & Czernik
1999]. In phase separation, the heavy mainly lignin-derived fraction separates
out from the aqueous fraction. This type of phase separation occurs for example
in fresh liquid, when the moisture content of the feedstock has been too high
(>15 wt%).
44
Table 6. Compound classes in pyrolysis liquids (D. Radlein in Bridgwater et al.
1999).
Compound class
C1 compound (formic acid, methanol
and formaldehyde, CO2)
C2–C4 linear hydroxyl and oxo substituted aldehydes and ketones
Hydroxyl, hydroxymethyl and/or oxo
substituted furans, furanones and pyranones (C5–C6)
Anhydrosugars, incl. anhydrooligosaccharides (C6)
Water-soluble carbohydrate derived
oligomeric and polymeric material
Monomeric methoxyl substituted phenols
Pyrolytic lignin
Composition range Hydrophilicity
(wt% of organic (arbitary scale 4
is highest)
fraction of pyrolysis liquid
4
5–10
15–35
4
10–20
3
6–10
4
5–10
4
6–15
2
15–30
1
Pyrolysis liquids differ to some extent in their ability to dissolve water [Oasmaa
et al. 1997]. A typical phase diagram for a pyrolysis liquid – water system is
shown in Figure 9. The weight percent of organics (or water) in each phase is
presented as a function of the global organic material concentration in the liquid.
The segment of the diagonal in the upper right corner corresponds to a singlephase liquid. With the addition of water to this particular pyrolysis liquid, the
phase separation occurs at 25 wt% of water (75 wt% of organics) in the liquid.
The bottom layer, which is represented by the upper branch in the diagram, contains more organic, mostly lignin-derived compounds, while the upper layer
(lower branch) is the aqueous fraction, with mostly carbohydrate-derived components. For example, in the overall water content of 50% the aqueous fraction
would contain 27% of organics and 73% of water, while the organic phase
would contain 72% of organics and 28% of water [Oasmaa & Czernik 1999].
45
Water in System [wt%]
100
80
60
40
20
0
0
100
Single Phase
Liquid
Organic Phase
(expected)
80
Organics in
Phase [wt%]
20
Water in
Phase [wt%]
Organic Phase
40
60
40
60
Aqueous
Phase
20
80
0
0
20
40
60
80
100
100
Organics in System [wt%]
Figure 9. Pyrolysis liquid - water phase diagram [Peacocke et al. 1994].
By adding increasing amounts of water to pyrolysis liquids, a phase separation
can be forced to occur (Appendix C/1). When adding excess water, the waterinsoluble lignin-derived fraction separates out of the aqueous phase (Appendix
G). The amount of water insolubles varies for different pyrolysis liquids: 34–35
wt% (dry basis) for hardwood, 32–35 wt% (dry basis) for softwood (pine) and
forest residue (pine/spruce), and 17–28.5 wt% (dry basis) for straw liquids. Extractives are included to water insolubles [Oasmaa & Kuoppala 2002b].
7.2.2 Solubility in other solvents
The solubility of pyrolysis liquids in solvents other than water is significantly
affected by the degree of polarity. Good solvents for highly polar white wood
and straw pyrolysis liquids are alcohols, like ethanol and methanol. These solvents dissolve practically the whole pyrolysis liquid excluding solids (char).
Acetone is also a good solvent for white wood liquids, but causes a precipitation/reaction in straw liquid (see Chapter 5.2).
46
White wood pyrolysis liquids do not dissolve in hydrocarbons like hexane, diesel fuels and polyolefins. However, forest residue and bark liquids contain extractives which are soluble in n-hexane. For dissolving forest residue liquids a
mixture of a polar (alcohol) and a neutral (dichloromethane, dioxane, hexane)
solvent is needed (see 5.2 and Appendix D).
For cleaning-up and washings basic solvents, like NaOH (Sodium Hydroxide)
and machine wash agents, are efficient in laboratory.
7.3 Stability
Pyrolysis liquids are not stable like conventional petroleum fuels due to their
high amount of reactive oxygen containing compounds and low-boiling volatiles. A comprehensive overview on stability of pyrolysis liquids is given by
Diebold in Bridgwater et al. 2001. The instability of pyrolysis liquids can be
seen as:
•
•
•
Evaporation of volatile components under air. Possible reactions with air.
Slow increase in viscosity and phase-separation ("aging").
Fast increase in viscosity/polymerisation and phase-separation by heating.
Exposure to air must be prevented for several reasons, e.g., loss of light compounds causing strong odours in the environment and oxygen causes polymerisation reactions with consequent sedimentation of heavy compounds. Hence,
storage, handling and transport must be performed in air-tight, non-ventilated
containers and transfer from the storage tank to a transport vehicle and reverse
must be carried out with a simultaneous exchange of gas to prevent air inlet and
exhaust to ambient air [Oasmaa et al. 1997, Hägerstedt & Jakobsson 1999].
Pyrolysis liquids contain compounds that, during storage or handling at ambient
temperatures, can react with themselves to form larger molecules. The main
chemical reactions observed are polymerisation of double-bonded compounds
[Polk & Phingbodhippakkiya 1981], as well as etherification and esterification
occurring between hydroxyl, carbonyl, and carboxyl group components [Czernik
et al. 1994], in which water is formed as a byproduct. These reactions result in
undesirable changes in physical properties, such as increase of viscosity, mo-
47
Viscosity @40°C, cSt
lecular weight and water content with a corresponding decrease of volatility
[Oasmaa & Czernik 1999]. Aging reactions are fastest in a fresh pyrolysis liquid
(Figure 10) and retard in time (Figure 10). Due to the instability of the pyrolysis
liquids, special care has to be taken in handling, transporting, storing, and using
the liquids.
24
23
22
21
20
0
1 2
3 4
5
6 7
8 9 10 11 12 13 14 15
Time, days
Viscosity @40°C, cSt
Figure 10. Viscosity increase of a forest residue liquid (bottom phase) during
the first two weeks after production.
40
35
30
25
20
0
1
2
3
4
5
6
7
8
9 10 11 12 13
Time, months
Figure 11. Viscosity increase of a forest residue liquid (bottom phase) during
one year after production.
48
When heating pyrolysis liquid four stages are observed:
1.
Thickening. The viscosity of the liquid increases because of mainly polymerisation reactions. At ambient temperatures the viscosity of a typical pyrolysis liquid roughly doubles in a year. The increase in viscosity is faster at
elevated temperatures (Figure 12). The viscosity change is different for
various pyrolysis liquids.
2.
Phase separation. Water is formed as by-product in aging reactions (etherification, esterification, condensation reactions). An aqueous phase separates
out of the heavy lignin-rich phase, when the total water content of the liquid
exceeds about 30 wt% (see Chapter 7.2).
3.
Gummy formation from the heavy lignin-rich-phase, if the temperature is
raised (abt. 140°C).
4.
Char/coke formation from the gummy phase at higher temperatures.
Viscosity @50°C, mPas
300
250
20 °C
200
35 °C
50 °C
150
80 °C
100
Linear (20 °C)
50
Linear (35 °C)
0
0
20
40
60
80
100
Time, days
Figure 12. Viscosity change of a hardwood pyrolysis liquid at 20–80°C [Oasmaa
et al. 1997].
49
Because of this thermal instability, heating should be done indirectly with a lowtemperature surface, e.g., warm water heat exchanger, jacketed tanks. Temperatures below 50°C are recommended for storage and pumping.
A simple test (Appendix H/1) was developed for a quick comparison of the stability of different pyrolysis liquids [Oasmaa et al. 1997]. In the test, the pyrolysis
liquid is kept at fixed temperature for a certain time, and the increase in viscosity
is measured. The presence of air (10 vol%) increases only slightly the viscosity
(Appendix H/2), and hence, for practical reasons, it is recommended to perform
the test without any nitrogen-purge with a fixed (10 vol%) free-space volume.
This test has been included to the round robin 2000 [Bridgwater et al. 2001] and
the results showed that more precise instructions and more testing for stability
test are needed. Other test conditions used were 6 hours at 80°C and 1 week at
40°C. The stability test is recommended to be used for internal comparison of
pyrolysis liquids owing similar initial viscosities.
For a pine liquid (water content 21 wt%, viscosity @40°C 35 mPas) the
viscosity increase (95%) in the test conditions 24 hours at 80°C correlated to the
viscosity increase in one year storage at room temperature. Addition of an
organic solvent, e.g. methanol, ethanol, or isopropanol, improves the stability of
the pyrolysis liquid (Diebold & Czernik 1997, Oasmaa et al. 1997). For
example, addition of 2, 5, and 10 wt% ethanol to the pine pyrolysis liquid
mentioned above the viscosity increase decreased to 56, 48, and 15%,
respectively. At higher temperatures, the ethanol addition had also a favourable
effect on viscosity (Oasmaa et al. 1997).
For forest residue liquid (bottom phase) the correlation (stability test compared
to room temperature storage) has been similar. Because of higher water content
of forest residue liquid test conditions 24 hours at 80°C has often yielded to
phase-separation and the viscosity of the liquid could not be determined for aged
sample. Hence, test conditions 1 week at 40°C has been used for forest residue
liquid. At these conditions the viscosity increase correlates to three months
storage at room temperature.
50
8. Fuel oil specifications
At present, there are no national or internationally recognised standard of fuel
specifications for biomass-derived fast pyrolysis liquids. Progress is being made
in this area, but due to the lack of a wide availability of consistent pyrolysis liquids, coupled with ongoing developments in the methods for fast pyrolysis liquids, standards are not expected for another 1–3 years. Commercial application
of biomass pyrolysis liquids requires an adequate description of their relevant
properties. The specifications for standard fuel oils have been established by
ASTM and similar organizations in respective countries. They define property
ranges for different classes of fuels marketed for different specialized applications [Oasmaa & Czernik 1999].
The significance of the specification standards relate to their development for
petroleum-derived fuel applications. When considering the applications of these
standards to a non-petroleum-derived fuel, one must understand the basis for the
standard and justify the appropriateness of its application [Elliott 1983]. Elliott
assessed the specification standards for various pyrolysis liquids in IEA BLTF
(International Energy Agency Biomass Liquefaction Test Facility) project. The
classification was based on ASTM standards D-396 for fuel oils, D-975 for diesel fuels, and D-2880 for gas turbine fuels. The IEA PYRA group suggested
[Diebold et al. 1997] specifications shown in Table 7. Feedback from pyrolysis
liquid end-users was requested.
Based on the diesel engine and boiler tests performed, an engine manufacturer,
Wärtsilä NSD Oy, and a potential bio-oil user, Birka Energi (former Stockholm
Energi AB), set the requirements for heating values and water, solids, and ash
content that are presented in Table 8 [Oasmaa & Czernik 1999]. For combustion,
tighter specifications, less property variation, has recently been acknowledged
[Oasmaa & Kytö 2000, Oasmaa & Meier in Bridgwater 2001].
It is necessary to conduct large-scale and long-duration tests on boilers, engines
and turbines before accepting pyrolysis liquids as commercial fuels. These tests
will require a substantial supply of liquids with specified properties established
by the producers and users.
51
Table 7. Proposed specifications for biomass pyrolysis liquids [Diebold et al.
1997].
Viscosity, cSt
Ash, wt%
Pour point, °C min
CCR, wt%
Max. 0.1 µm filtered
ethanol insolubles
Accelerated aging
rate @ 90 °C cSt/h
Water, wt% of
wet oil, max.
LHV, MJ/L min.,
wet oil
C, wt% dry
H, wt% dry
O, wt% dry
S, wt% dry
N, wt% dry
K + Na, ppm
Phase Stability
@ 20 °C 8 h
@ 90 °C
Flash point, °C min.
Density, kg/m3
Light
bio-oil
(~ASTM #2)
1.9-3.4 FO
1.9-4.1 D
1.9-4.1 GT
@ 40 °C
0.05 FO
0.01 D
0.01 GT
report
report
Light-Medium
bio-oil
(~ASTM #4)
5.5-24
@ 40 °C
Medium
bio-oil
(~PORL100)
17-100
@ 50 °C
Heavy
bio-oil
(~Can. #6)
100-638
@ 50 °C
0.05 FO
0.10 D
0.10 FO
0.10 FO
report
report
report
report
report
report
0.01 FO
0.05
0.10
0.25
report
report
report
report
32
32
32
32
18
report
"
"
0.1 max.
0.2 max.
report
0.5 GT
single
phase
18
report
"
"
0.1 max.
0.2 max.
report
18
report
"
"
0.2 max.
0.3 max.
report
report
report
"
"
0.4 max.
0.4 max.
report
single
phase
single
phase
single
phase
52
report
55
report
60
report
60
report
52
Table 8. Specifications for biomass pyrolysis oil for diesel engines and boilers.
Diesel engine tests by Wärtsilä
Specifications
Homogeneity
Water, wt%
7-day storage
max 26
HHV, MJ/kg
LHV, MJ/kg
Ash, wt%
Solids, wt%
<50 µm
<25 µm
min 18
min 16
max 0.1
max 1
100
min 90
Allowable variation
(10%) for a container
(surface/bottom)
no phase-separation
max 10% difference,
max 26
max 10% difference,
min 18 and 16
0–0.1
0–1
-
53
Tests at Birka
Combustion
Specifications
max 25
min 19
min 17
max 0.1
max 1
-
9. Recommendations
Pyrolysis liquids exhibit unusual properties, which are not apparent in conventional hydrocarbon liquids. The aim of this guide is to highlight the difficulties
with the standard methods for property determination and to elucidate and propose
modified standard and new test methods as appropriate, based on the current state
of knowledge. Further developments in the test methods will continue, however,
this guide provides basic data for the producer, tester and end-user of the liquids.
Based on the wide range of properties assessed and evaluated in this guide, the
following modifications to the standard methods are recommended:
Table 9. Analytical methods for wood-based pyrolysis liquids.
Analysis
Method
Comment
Water, wt%
Solids, wt%
ASTM E 203
Ethanol insolubles
Methanol-Dichloromethane
insolubles
Particle size distribution Microscopy+Particle Counter
Conradson carbon, wt% ASTM D 189
Ash, wt%
EN 7
CHN, wt%
ASTM D 5291
Sulphur and chlorine, wt% Capillary electrophoresis
Alkali metals, wt%
AAS
Metals, wt%
ICP, AAS
Density (15°C), kg/dm3 ASTM D 4052
Viscosity (20, 40°C), cSt ASTM D 445
Viscosity, mPas
Rotational viscometry
Pour point, °C
ASTM D 97
Heating value, MJ/kg
calorimetric (HHV)
DIN 51900
effective (LHV)
Flash point, °C
ASTM D 93
pH
pH meter
Water insolubles, wt%
Water addition
Stability
80°C 24 hours
40°C 1 week
1
2
2
Sample
size
1g
30 g
30 g
3
4
5
6
7
8
9
10
11
12
13
1g
2–4 g
40 ml
1 ml
2–10 ml
50 ml
50 ml
4 ml
80 ml
40 ml
80 ml
14
1 ml
15
16
17
18
18
150 ml
50 ml
5 ml
200 ml
200 ml
Sample size = Minimum amount of pyrolysis liquid needed for carrying out the analysis
including duplicates.
54
Comments to Table 9.
Note 1: When handling reagents in the procedures described in the following
appendices, please observe all relevant health and safety procedures (Appendix A)
applicable.
Note 2: Extractive-rich samples, like forest residue liquids, foam easily. A gentle
mixing, like rolling the sample bottle carefully, is recommended. See Appendix B.
1
Karl-Fischer titration. Methanol-Chloroform (3:1) as a solvent. Water addition method for calibration. HYDRANAL K reagents (Composite 5K and
Working Medium K) in case of a fading titration end-point. 50 ml solvent
for two determinations. Sample size about 0.25 g (water content >20 wt%).
Stabilisation time 30 s. See appendix C.
2
Millipore or multi-place filtration system, 1 µm filter, sample size 1–15 g in
order to obtain 10-20 mg residue, sample:solvent = 1:100, solvent: ethanol
for white wood liquids, methanol-dichloromethane for forest residue liquids
(Appendix D).
3
Microscopy with photo analysis programme (Appendix E).
4
Controlled evaporation of water to avoid foaming. The implication of carbon residue is unclear.
5
Controlled evaporation of water to avoid foaming.
6
Proper homogenisation. For forest residue liquids careful rolling of the
sample bottle. Large sample size as possible. Triplicates.
7
Sample pretreatment by combustion according to ASTM D 4208
8
Wet combustion as a pretreatment method. See appendix F.
9
Wet combustion (Appendix F) as a pretreatment method. In samples with a
high amount of silicates, silicon can precipitate as SiO2 during the sample
pretreatment. This may yield an error in silicon. For accurate determination of Si the sample should be ashed by dry combustion and a fusion cake
prepared from the ash.
10 Careful mixing of foam-prone forest residue liquids in order to avoid air
bubbles.
11 Cannon-Fenske viscometer tubes at room temperature and for nontransparent liquids, Ubbelohde tubes may be used for transparent liquids.
55
No prefiltration of the sample if visually homogenous. Elimination of air
bubbles before sampling. Equilibration time 15 minutes.
12 Precise temperature measurement. Cover to the sample holder above 40°C.
13 No preheating of the sample.
14 Use of a fine cotton thread for ignition. Lower heating value (LHV) obtained from calorimetric heating value and hydrogen analysis.
15 Elimination of air bubbles before sampling.
16 Frequent calibration of the pH meter.
17 Addition of 5 g pyrolysis liquid into water, see Appendix G
18 90 ml (45 ml) pyrolysis liquid in 100 ml (50 ml) tight glass bottles, heating
in a heating oven (Appendix H). Measuring of increase in viscosity and water. Viscosity determination at 40°C according to ASTM D 445.
Other analyses from literature, which were not tested within this study, are presented in Table 10. Properties of pyrolysis liquids are summarised in Table 11.
Table 10. Analytical methods for wood-based pyrolysis liquids. Literature data
[Bridgwater et al. 1999].
Property
Density
Thermal conductivity
Specific heat capacity
Setting point
Boiling curve
Coke residue
Ignition limit
Vapour pressure
Surface tension
Sulphur
Standard method
ASTM D941
ASTM D1298-85
No standard
No standard
DIN51583
ASTM D86-82
ASTM D524-88
DIN51603
IP69/89
ASTM D 971-50
ASTM D4208
56
Suitability of the method
Can be used
Can be used
Rough estimate
Rough estimate
Can be used
Cannot be used
Can be used
Limited testing
Can be used
Limited testing
Can be used
Table 11. Properties of pyrolysis liquids.
Property
Density (15°C), kg/dm3
Lower heating value, MJ/kg
Viscosity-kinematic, cSt
Thermal conductivity, W/mK
Specific heat capacity, J/gK
Pour point
Coke residue, wt%
Flash point, °C
Ignition limit, °C
Ignition temperature, °C
Water, wt%
Char, wt%
Vapour pressure, kPa
Surface tension, mN/m
Carbon, wt% (dry)
Hydrogen
Nitrogen
Oxygen
S, ppm
Cl, ppm
Ash, wt%
pH
K+Na, ppm
Range (wet basis)
1.11–1.30
13–18
10–80 @50°C
0.35–0.43
2.6–3.8 @ 25–60°C
9–-36
14–23
40–110
110–120
600–700
20–35
0.01–1
5.2 @33.5°C
29.2
32–49 (48–60)
6.9–8.6 (5.9–7.2)
0–0.4
44–60 (34–45)
60–500
3–75
0.01–0.20
2.0–3.7
5–500
The use of biomass-derived pyrolysis liquids in heating and electricity generation applications is increasing significantly. For the producer, a ready reference
for the determination of properties and how to handle the liquids is essential. It is
hoped that this guide will allow those active in research, development and commercial applications to better understand how to evaluate the properties of the
liquids, relevant to the application and allow the liquids to be used in a safe and
environmentally compliant way.
The use of the modified methods [derived from conventional fuel oil methods]
proposed in this guide is aimed to avoid some of the more common problems
associated with the liquids and improve their usability in a range of applications.
57
References
Agblevor, F.A., Besler, S. & Wiselogel, A. E. 1996. Production of oxygenated
fuels from biomass: impact of storage conditions. Fuel Science and Technology
International, vol. 14, no. 4, pp. 589–612.
Anon. 2000. United Nations Recommendations on the Transport of Dangerous
Goods Model Regulations. 11th Revised Edition. ISBN 92-1-139067-2.
ASTM D 95. 1999. Water in petroleum products by distillation. Easton, MD:
American Society for Testing and Materials.
ASTM D 93. 1991. Standard test method for flash point by PenskyMartens
closed tester. Easton, MD: American Society for Testing and Materials.
ASTM D 97. 1993. Standard test method for pour point of petroleum oils.
Easton, MD: American Society for Testing and Materials.
ASTM D 95. 1988. Standard test method for water in petroleum products and
bituminous materials by distillation. Easton, MD: American Society for Testing
and Materials.
ASTM D 130. 1994. Standard test method for detection of copper corrosion
from petroleum products by the copper strip tarnish test. Easton, MD: American
Society for Testing and Materials.
ASTM D 189. 1988. Standard test method for Conradson carbon residue of petroleum products. Easton, MD: American Society for Testing and Materials.
ASTM E 203. 1996. Standard test method to water using volumetric Karl
Fischer Titration. Eaton, MD: American Society for Testing and Materials.
ASTM D 445. 1988. Standard test method for kinematic viscosity of transparent
and opaque liquids (and the calculation of dynamic viscosity). Easton, MD:
American Society for Testing and Materials.
58
ASTM D 613. 1995. Standard test method for cetane number of diesel fuel oil.
Easton, MD: American Society for Testing and Materials.
ASTM D 665. 1995. Standard test method for rust preventing characteristics of
inhibited mineral oil in the presence of water. Easton, MD: American Society for
Testing and Materials.
ASTM D 1744. 1988. Standard test method for water in liquid petroleum products by Karl Fischer reagent. Easton, MD: American Society for Testing and
Materials.
ASTM D 2500. 1988. Standard test method for cloud point of petroleum oils.
Easton, MD: American Society for Testing and Materials.
ASTMD 2699. 1969. Knock characteristics of motor fuels by the research
method. Easton, MD: American Society for Testing and Materials.
ASTM D 2783. 1993. Standard test method for measurement of extreme pressure properties of lubricating fluids (fourball method). Easton, MD: American
Society for Testing and Materials.
ASTM D 4052. 1988. Standard test method for density and relative density of
liquids by digital density meter. Easton, MD: American Society for Testing and
Materials.
ASTM D 4208. 1988. Standard test method for total chorine in coal by the oxygen bomb combustion/ion selective electrode method. Easton, MD: American
Society for Testing and Materials.
ASTM D 4239. 1991. Standard test methods for sulfur in the analysis sample of
coal and coke using high temperature tube furnace combustion methods. Philadelphia, PA: American Society for Testing and Materials.
ASTM D 5291. 1992. Standard test methods for instrumental determination of
carbon, hydrogen, and nitrogen in petroleum products and lubricants. Easton,
MD: American Society for Testing and Materials.
59
Aubin, H. & Roy, C. 1980. Study on the corrosiveness of wood pyrolysis oils.
Fuel Science & Technology International, vol. 8, pp. 77–86.
Bertoli et al. JOR3-CT98-0253, Final report.
Bridgwater, A. V. (ed.). 2001. Progress in Thermochemical Biomass Conversion. Oxford: Blackwell Science.
Bridgwater, A., Czernik, S., Diebold, J., Meier, D., Oasmaa, A., Peacocke, G.,
Piskorz, J. & Radlein, D. 1999. Fast pyrolysis of biomass: A handbook. IEA
Bioenergy Task XIII Pyrolysis Activity. Newbury: CPL Press. 188 p. ISBN 1872691-07-2.
Bridgwater, A. et al. Fast pyrolysis of biomass. Handbook 2. To be published in
2001.
Bridgwater, A. V. 1995. Engineering developments in flash pyrolysis technology. In: Proc. Conference on Bio-oil Production and Utilisation, Estes Park, CO,
24–26 Sept. 1994. Golden, CO: NREL.
Chum, H. L. & McKinley, J. 1988. Report on characterization of biomass pyrolysis liquid products. In: Bridgwater, A. V. & Kuester, J. L. (eds.). Research in
thermo chemical biomass conversion, Phoenix, Arizona, April 1988. New York:
Elsevier Appl. Sci. Pp. 1177–1180.
Czernik, S., Johnson, D. K. & Black, S. 1994. Stability of wood pyrolysis liquid.
Biomass and Bioenergy, vol. 7, no. 1–6, pp. 187–192.
Czernik, S. 1994. Storage of biomass pyrolysis oils. In: Proc. Specialist Workshop on Biomass Pyrolysis Oil Properties and Combustion, Estes Park, CO, 26–
28 Sept. 1994. Golden, CO: NREL. Pp. 67–76. CP-430-7215.
Diebold, J. P. & Bridgwater, A. V. 1997. Overview of fast pyrolysis of biomass
for the production of liquid fuels. In: Bridgwater, A. V. & Boocock, D. G. B.
(eds.). Developments in thermochemical biomass conversion. London: Blackie
Academic & Professional. Pp. 5–26.
60
Diebold, J. P. & Czernik, S. 1997. Additives to lower and stabilize the viscosity
of pyrolysis oils during storage. Energy & Fuels, vol. 11, pp. 1081–1091.
Diebold, J. P., Milne, T., Czernik, S., Oasmaa, A., Bridgwater, A..V., Cuevas,
A., Gust, S., Huffman, D. & Piskorz, J. 1997. Proposed specifications for various grades of pyrolysis oils. In: Bridgwater, A.V. & Boocock, D. G. B. (eds.).
Developments in Thermochemical Biomass Conversion, Banff, 20–24 May 1996.
Glasgow: Blackie Academic & Professional. Vol. 1. Pp. 433–447.
Diebold, J. P., Scahill, J. W., Czernik, S., Phillips, S. D. & Feik, C. J. 1997. Progress in the production of hotgas filtered biocrude oil at NREL. In: Bridgwater,
A. V. & Hogan, E. N. (eds.). Proc. 2nd EU Canada Workshop on Thermal Biomass Processing. Newbury: CPL Scientific Information Services Ltd. Pp. 66–81.
Diebold, J. P. 1999. A review of the toxicity of biomass pyrolysis liquids formed
at low temperatures. In: Bridgwater, A., Czernik, S., Diebold, J., Meier, D.,
Oasmaa, A., Peacocke, C., Piskorz, J. & Radlein, D. (eds.). Fast pyrolysis of
biomass: A handbook. Newbury: CPL Press. Pp. 135–163.
DIN EN 7. 1975. Determination of ash from petroleum products. Berlin: DIN
Deutsches Institut für Normung e.V.
DIN 51900. 1977. Testing of solid and liquid fuels; determination of gross calorific value by the bomb calorimeter and calculation of net calorific value;
method with the adiabatic jacket. Berlin: DIN Deutsches Institut für Normung
e.V.
Dyroff, G. V. 1993. Manual on significance of tests for petroleum products. 6.
ed. Philadelphia: ASTM.
Elliott, D. C. 1983. Analysis and upgrading of biomass liquefaction products.
Final report. Vol. 4. IEA Cooperative project D1 Biomass Liquefaction Test
Facility Project. Richland, Washington: Pacific Northwest Laboratory. 87 p. +
app.
61
Fagernäs, L. 1995. Chemical and physical characterisation of biomass-based
pyrolysis oils. Literature review. Espoo: Technical Research Centre of Finland.
113 p. + app. 2 p. (VTT Research Notes 1706.)
Fuleki, D. 1999. Bio-fuel system material testing, PyNE Newsletter, Issue no. 7,
April 1999. Birmingham, UK: Aston University, Bio-Energy Research Group.
Gust, S. 1997. Combustion experiences of flash pyrolysis fuel in intermediate
size boilers. In: Bridgwater, A.V & Boocock, D.G.B. Developments in thermochemical biomass conversion. Volume 1. Pp. 481–488.
Hägerstedt, L.-E. & Jakobsson, A. 1999. Bio fuel oil for power plants and boilers. Final report: Handling, storage and transport of bio fuel oil. Stockholm:
Birka Teknik och Miljö AB. JOR3-CT95-0025.
Hallgren, B. 1996. Test report of Metlab Miljö AB. Skelleftehamn: Metlab Miljö
AB. 17 p. (Reg. no. ALL1668, 1996 02 0809.)
Ikura, M. et al. 1998. JOR3-CT98-0253. Final report.
Jamieson, D. T., Irving, J. B. & Tudhope, J. S. The thermal conductivity of petroleum products: a survey to 1974. National Engineering Laboratory Liquid
Thermal Conductivity Survey. Glasgow: HM Stationary Office, 1975.
Jay, D. C., Rantanen, O., Sipilä, K. & Nylund, N.-O. 1995. Wood pyrolysis oil for
diesel engines. In: Proc. 1995 Fall Technical Conference, Milwaukee, Wisconsin,
24–27 Sept 1995. New York: ASME, Internal Combustion Engine Division.
Leroy, J., Choplin, L. & Kallaguime, S. 1988. Rheolological characterization of
pyrolytic wood derived oils: Existence of a compensation effect. Chem. Eng.
Comm., vol. 71, pp. 157–176.
Lewis, C. W., Baumgardner, R. E., Stevens, R. K., Claxton, L. D. & Lewtas, J.
1988. Contribution of wood smoke and motor vehicle emissions to ambient aerosol mutagenicity. Environment Science and Technology, vol. 22, pp. 968–971.
62
McKinley, J. W., Overend, R. P. & Elliott, D. C. 1994. The ultimate analysis of
biomass liquefaction products: The results of the IEA round robin #1. In: Proc.
Biomass pyrolysis oil properties and combustion meeting, 26–28 September
1994, Estes Park, CO. Golden, CO: NREL. Pp. 34–53.(NRELCP4307215.)
McKinley, J. W. 1989. Biomass liquefaction: centralized analysis. Final report.
Vancouver: B. C. Research. (Project No. 403837. DSS File No. 2321646192.)
Meier, D., Oasmaa, A. & Peacocke, G. V. C. 1997. Properties of fast pyrolysis
liquids: status of test methods. Characterisation of fast pyrolysis liquids. In:
Bridgwater, A. V. & Boocock, D. G. B. (eds.). Developments in thermochemical
biomass conversion, Banff, 20–24 May 1996. London: Blackie Academic & Professional. Vol. 1. Pp. 391–408.
Milne, T. A., Brennan, A. H. & Glenn, B. H. 1990. Sourcebook of methods of
analysis for biomass and biomass conversion processes. London: Elsevier Appl.
Sci. 327 p. + app.
Oasmaa, A. & Kytö, M. 2000. Pyrolysis oil combustion tests in an industrial
boiler. Pyne Pyrolysis Network, no. 10, pp. 13
Oasmaa, A. & Czernik, S. 1999. Fuel oil quality of biomass pyrolysis oils. In:
Biomass. A growth opportunity in green energy and value-added products. Vol.
2. Kidlington: Elsevier Science. Pp. 1247–1252.
Oasmaa, A. & Kuoppala, E. Composition and properties of forest residue liquid.
2002b. To be submitted to Energy & Fuels.
Oasmaa, A., Leppämäki, E., Koponen, P., Levander, J. & Tapola, E. 1997.
Physical characterisation of biomass-based pyrolysis liquids. Application of
standard fuel oil analyses. Espoo, VTT. 46 p. + app. 30 p. (VTT Publications
306.) ISBN 951-38-5051-X
Peacocke, G. V. C. & Bridgwater, A. V. 2001. ''Transport, handling and storage
of biomass derived fast pyrolysis liquid'', In: Bridgewater, A. V. (ed.). Progress
in Thermochemical Biomass Conversion, Vol. 2, p. 1482–1499.
63
Peacocke, G. V. C., Russell, P. A., Jenkins, J. D. & Bridgwater, A. V. 1994.
Physical properties of flash pyrolysis liquids. Biomass & Bioenergy, vol. 7, no. 1
6, pp. 169–177.
Polar, J. P. (ed.). 1961. A guide to corrosion resistance, Caxton Press, p. 190.
Polk, M. B. & Phingbodhippakkiya, M. 1981. Development of methods for the
stabilization of pyrolytic oils. Cincinnati, Ohio: EPA. (EPA-600/2-81-201.)
Putnam, K. P., Bombick, D. W., Avalos, J. T. & Doolittle, D. J. 1999. Comparison of the cytotoxic and mutagenic potential of liquid smoke food flavourings,
cigarette smoke condensate and wood smoke condensate. Food and Chemical
Toxicology, vol. 37, pp. 1113–1118.
Radovanovic, M., Venderbosch, R. H., Prins, W. & van Swaaij, W. P. M. 2000.
Some remarks on the viscosity measurement of pyrolysis liquids. Biomass and
Bioenergy, vol. 18, pp. 209–222.
Rick, F. & Vix, U. 1991. Product standards for pyrolysis products for use as fuel
in industrial firing plants. In: Bridgwater, A. V. & Grassi, G. (eds.). Biomass
pyrolysis liquids upgrading and utilization. London & New York: Elsevier Applied Science. Pp. 177–218.
Riedel-de Haën. 1995. HYDRANAL Manual, Eugen Scholz reagents for Karl
Fischer titration. Seelze. P. 123.
Shihadeh, A. L. 1998. Rural electrification from local resources: biomass pyrolysis oil combustion in a direct injection diesel engine. PhD Thesis at the Massachusetts Institute of Technology. P. 34.
Sipilä, K., Kuoppala, E., Fagernäs, L. & Oasmaa, A. 1998. Characterisation of
biomass-based flash pyrolysis oils, Biomass and Bioenergy, vol. 14, no. 2, pp.
103–113.
Solantausta, Y., Nylund, NO., Westerholm, M., Koljonen, T. & Oasmaa, A.
1993. Wood pyrolysis liquid as fuel in a diesel power plant. Bioresource Technology, vol. 46, no. 1 & 2, pp. 177–188.
64
Soltes, E.J. & Lin, J.-C. K. 1984. Hydroprocessing of biomass tars for liquid
engine fuels. In: Tillman, D. A. & Jahn, E. C. (eds.). Progress in Biomass Conversion. New York: Academic Press. Pp. 1–69.
Suppes, G. J., Natarajan, V. P. & Chen, Z. 1996. Autoignition of select oxygenated fuels in a simulated diesel engine environment. In: AICHE 1996 National
Meeting, 26 Febr. 1996. Paper 74e. New Orleans, LA: AICHE. 9 p. (Available
from the Engineering Library, New York City.)
65
LIST OF APPENDICES
Appendix A:
MSDS – Material health and safety data
sheet for pyrolysis
Appendix B:
Homogenisation and sampling
Appendix C:
Determination of water content
Appendix D:
Determination of solids content
Appendix E:
Particle size distribution for pyrolysis
liquid
Appendix F:
Alkali metals and metals
Appendix G:
Water extraction method for pyrolysis
liquids
Appendix H:
Stability test method
Where appropriate for methodologies, looked at: Oasmaa, A. Leppämäki, E.,
Koponen, P., Levander, J. & Tapola, E. Physical characterisation of biomassbased pyrolysis liquids. Application of standard fuel oil analyses. Espoo: VTT
Energy, 1997. 46 p. + app. 30 p. (VTT Publications 306.)
http://www.inf.vtt.fi/pdf/publications/1997/P306.pdf
Note: When handling reagents in the procedures described in the following
appendices, please observe all relevant health and safety procedures
applicable.
Appendix A:
MSDS – Material health and safety data
sheet for pyrolysis liquids
Prepared by Dr.C.Peacocke and Dr.A.V.Bridgwater, Aston University, UK.
Modifications to Chapters 2.0 and 10.0 based on IEA PyNe MSDS [Bridgwater
et al. 1999].
NOTE:
There is at present no official recognition of the information contained within
this document. The Health and Safety Executive do not have an official designation for pyrolysis liquids.
This MSDS has no legal validity and represents a consensus view of those active
in biomass fast pyrolysis, that using publicly available information represents the
best effort that can be made at this time.
1.0
Chemical Identification
Pyrolysis liquid (also known as pyrolysis oil, bio-crude-oil, bio-oil bio-fuel-oil,
pyroligneous tar, pyroligneous acid, wood liquids, wood oil, liquid smoke, wood
distillates)
2.0
Composition/Information on Ingredients
Complex mixture of highly oxygenated hydrocarbons. A mixture of three to four
hundred chemicals derived by the thermal decomposition of biomass. A typical
pyrolysis oil may be composed as follows:
Composition
organic acids
aldehydes and hydroxyaldehydes
A1
wt%
5–10
5–20
ketones and hydroxyketones
anhydrosugars
degraded lignin, phenolics
hydrocarbons
water
0–10
5–30
20–30
0–5
15–35
A list of components which may be found in pyrolysis oils is appended (Appendix 1).
3.0
Synonyms
Bio-oil, pyrolysis liquid, pyroligneous oil, bio crude oil, pyroligneous liquid,
liquid wood, bio-fuel oil, wood tar, wood oil.
4.0
Hazards Identification
Label Precautionary Statements
•
•
•
•
•
•
•
•
•
Harmful by inhalation, in contact with skin and if swallowed.
Irritating to eyes, respiratory system and skin.
Corrosive (pH 1.5–3).
Only flammable with additional fuel source at 5wt% concentration such as
methanol, ethanol, acetone, or diesel and upon heating with suitable ignition source.
Possible mutagen, contains potentially carcinogenic compounds.
Keep container tightly closed in a cool well ventilated place.
Do not empty into drains.
Wear suitable gloves and eye/face protection when handling.
Avoid continuous exposure.
5.0
First Aid Measures
•
In case of contact with eyes flush with copious amounts of water for 15
minutes. Remove contaminated clothing.
In case of contact with skin flush with copious amounts of water. Remove
contaminated clothing.
If inhaled remove to fresh air. If breathing is difficult give oxygen. If not
breathing give artificial respiration.
•
•
A2
•
If swallowed, wash out mouth with water. Consume water to dilute. Call
doctor immediately.
6.0
Fire Fighting Measures
Extinguishing Media
Water, Carbon Dioxide, Foam, Powder
Special Fire Fighting Precautions
•
•
Wear self contained breathing apparatus and protective clothing to prevent
contact with eyes and skin. Do not inhale smoke from fire.
Use water spray to cool fire exposed containers.
7.0
Accidental Release Measures
7.1
Small Quantities (<100ml)
•
•
•
•
Wear rubber gloves and suitable eye and face protection.
Cover contaminated area with sawdust.
Take up sawdust and place in closed container.
Transport to approved landfill or incinerator.
7.2
Large Quantities (>100ml)
•
•
Evacuate area.
Wear rubber boots, rubber gloves, suitable eye/face protection and
NIOSH/MSHA approved respirator.
Cover contaminated area with sawdust. Take up sawdust and place in
closed container. Transport to approved landfill or incinerator.
•
8.0
Handling and Storage
•
•
Store in sealed container in darkness at room temperature.
Immiscible with water above 50% weight concentration. Soluble only in
solvents such as acetone or ethanol. Immiscible in hydrocarbon solvents.
Reacts with mild steel and impure copper due to high acidity. Attacks
buna rubber and in some cases causes rubber seals to swell.
•
A3
9.0
Exposure Controls/Personal Protection
•
Wear appropriate NIOSH/MSHA approved respirator, rubber gloves, rubber apron, and suitable eye/face protection.
Use in a well ventilated area or fume cupboard.
Safety shower and eye bath.
Do not breath vapour.
Wash thoroughly after handling.
•
•
•
•
10.0 Physical and Chemical Properties
Dark brown, viscous liquid with a smoky odour.
(All values are typical, not definitive, values)
pH
Density (kg/dm3)
Viscosity (cSt)
Pour point (°C)
Flash point (°C)
Boiling Curve
Autoignition Temperature
Upper Explosion Level:
Lower Explosion Level:
Vapour Pressure:
Vapour density:
2.5–3.5
1.2 @ 15°C
20–100 @ 40°C
20
50–70
Starts boiling below 100°C. Forms 20–50% of
solid residue.
High. Approximately above 500°C.
Unknown
Unknown
Similar to water
Unknown
11.0 Stability and Reactivity
Incompatibilities
Heat - 50% char formed upon continuous heating above 100°C
Hazardous Combustion or Decomposition Products.
Fumes of:
Carbon dioxide
Carbon monoxide
Light organics-2-propenal, acetic acid, formaldehyde, methanol, acetaldehyde
A4
12.0 Toxicological Information
•
•
•
•
•
Liquids produced at reactor temperatures greater than 600°C contain condensed polyaromatic ring compounds which have mutagenic effects.
Harmful if swallowed, inhaled, or absorbed through the skin.
Vapour is irritating to the eyes, mucous membranes and upper respiratory
tract.
Can sensitise skin.
Laboratory test have shown mutagenic properties.
13.0 Ecological Information
Very high BOD - no exact value yet
14.0 Disposal Considerations
Burn in a chemical incinerator observing all environmental regulations.
15.0 Transport Information.
Transport in sealed translucent container in cool conditions. Avoid prolonged
exposure to strong sunlight and heat.
Bibliography
•
•
•
•
•
Elliot, D. C. Comparative analysis of biomass pyrolysis condensates.
Rochland, Washington: Chemical Technology Department, Pacific Northwest Laboratory.
Longley, C. J., Howard, J. & Fung, D. 1994. Levoglucosan recovery from
cellulose and wood pyrolysis liquids., In: Bridgwater, A. V. (ed.). Advances
in thermochemical bBiomass conversion. London: Blackie. Pp. 1441–1451.
McKinley, J. 1989. Final report biomass liquefaction: centralized analysis.
Vancouver, B.C.: B.C. Research.
Material Safety Data Sheet prepared by G. Underwood for Red Arrow Products Inc.
Peacocke, G. V. C. of Aston University, Birmingham, England. Information
provided from personal experience.
A5
•
•
•
•
•
•
•
Piskorz, J., Radlein, D., Scott, D. S. & Czernik S. 1988. Liquid products
from the fast pyrolysis of eood and cellulose. In: Research in Thermochemical Biomass Conversion, p.557–571.
Piskorz, J., Scott, D.S., Radlein, D. & Czernik, S. 1992. New applications of
the Waterloo Fast Pyrolysis process. In: Hogan, E., Robert, J., Grassi, G. &
Bridgwater, A. (eds.). Biomass thermal processing, Pp. 64–73.
Rick, F. & Vix, U. 1988. Product dtandards for pyrolysis products for use as
fuel in industrial firing plants. In: Scott, D. S. & Fung, D. (eds.). Biomass
pyrolysis liquids upgrading and utilisation. Chemicals and fuels from biomass flash pyrolysis - part of the bioenergy development program (BDP).
Solantausta, Y, Diebold, J, Elliott, D.C, Bridgwater, A.V. & Beckman, D.
1993. Assessment of liquefaction and pyrolysis dystems. Espoo: Technical
Research Centre of Finland.
Scott, D. S., Piskorz, J. & Radlein, D. 1992. The yields of chemicals from
th
biomass based fast pyrolysis oils. In: Energy from Biomass and Wastes 16
ed.
Approved Carriage List-Information approved for the classification, packaging and labelling of dangerous goods for carriage by Road and Rail. Carriage of Dangerous Goods by Road and Rail (Classification, Packaging and
Labelling) Regulations 1994, (SI 1994/669), Health and Safety Commission,
HMSO ISBN 0 11 043669 5.
Approved methods for the classification and packaging of dangerous goods
for carriage by road and Rail. Carriage of Dangerous Goods by Road and
Rail (Classification, Packaging and Labelling) Regulations 1994. Health and
Safety Commission, HMSO, ISBN 0 11 041735 6.
A6
Appendix 1
Some Chemicals Identified in Pyrolysis Liquid.
Acids
oxopentanoic acid
acetic acid
benzoic acid
butyric acid
formic acid (methanoic acid)
glycolic acid
hexadecanoic acid
hexanoic acid
propanoic (propionic) acid
valeric acid (pentanoic acid)
Sugars
1,6-anhydroglucofuranose
cellobiosan
D-arabinose
D-glucose
D-xylose
fructose
oligosacharides
levoglucosan
Ketones
1-hydroxy 2-propanone
2,5-hexanedione
2-butanone
2-ethylcyclopentanone
2-methyl-2-cyclopenten-1-one
2-methylcyclohexanone
2-methylcyclopentanone
3-ethylcyclopentanone
3-methyl-2-cyclopenten-1-one
A7
3-methylcyclohexanone
3-methylcyclopentanone
3-methylindan-1-one
C8 (=) cyclic ketones
C8 cyclic ketones
C9 (=) cyclic ketones
C9 cyclic ketones
cyclohexanone
cyclopentanone
dimethylcyclopentanone
hydroxyacetone
methylcyclopentene-ol-one
trimethylcyclopentanone
Phenolics
(4-hydroxyphenyl)-1-ethanone
2,3,5-trimethylphenol
2,3,6-trimethylphenol
2,3-dimethyl phenol
2,4,6-trimethylphenol
2,4-xylenol
2,5,8-trimethyl-1-naphthol
2,5-dimethyl phenol
2,6-dimethoxy phenol
2-methoxy-4(1-propenyl)phenol
2-methoxy-4-n-propylphenol
3-methyl-1-naphthol
4-ethyl-1,3-benzenediol
4-ethyl-2-methoxyphenol
4-ethylguaiacol
4-hydroxy-3,5-dimethoxyphenylethanon
4-hydroxy-3-methoxybenzaldehyde (Vanillin)
4-methyl-2-methoxyphenol
4-propylguaiacol
6,7-dimethylnaphthol
C2 dihydroxybenzene
A8
C3 dihydroxybenzene
C3 phenols
C4 dihydroxybenzene
C4 phenol
C4 phenols
C5 (=) phenols
C5 dihydroxybenzene
C5 phenols
C6 (=) phenols
C6 dihydroxybenzene
C6 phenols
catechol
dimethyldihydroxybenzene
dimethylnaphthol
ethylmethylphenol
eugenol
guaiacol
hydroquinone
isoeugenol
m-cresol
methylbenzenediol
o-cresol
p-cresol
p-ethylphenol
phenol
resorcinol
syringaldehyde
trimethyldihydroxybenzene
trimethylnaphthol
Other Oxygenates
1-acetyloxy-2-propanone
2,3-dihydro-1,4-benzodioxin
2-butoxyethanol
2-furaldehyde
2-furanone
A9
2,5-dimethylfuran
2-methyl-2-butenal
2-methylcyclopentanol
2-methylfuran
3-hydroxy-2-methyl-4-pyrene
4-butyrolactone
5-hydroxymethyl-2-furanone
5-methyl furfural
5-methyl-2(3H)furanone
acetal
acetaldehyde
angelicalactone
acetol
dibenzofuran
ethylene glycol (ethan-1,2-diol)
formaldehyde
furfuryl alcohol
glyoxal (ethan-1,2-dial)
hydroxyacetaldehyde
methanol
methyl glyoxal
methyl formate
Hydrocarbons
1-eicosene
1-heneicosene
1-heptadecene
1-hexacosene
1-nonadecene
1-octadecene
1-pentacosene
1-tetracosene
1-tricosene
1-tridecene
2,6,10,14-tetramethylpentadecane
4-methylanisole
A10
benzene
dimethylcylopentene
eicosene
heneicosene
heptadecene
hexadecene
methylindane
nonadecene
octadecene
pentacosene
pentadecene
styrene
tetracosene
tetradecene
tetrahydrocyclopropyl(a)indene
toluene
tricosene
A11
Appendix B:
Homogenisation and sampling
B.1 Homogenisation
A small (containers <10 l) sample vessel is homogenised in a laboratory mixer
(about one hour at room temperature). Exractive-rich samples, like forest residue liquids, foam easily. Hence a gentle mixing, like rolling the sample bottle
carefully, is recommended.
Larger (10-100 l) containers can be turned upside down overnight, if the liquid is
old (> 6 months) and a sticky solids layer has settled at the bottom. The
pyrolysis liquid is mixed by a propeller mixer with top-mounted slow revolving
propeller stirrer. The pyrolysis liquid is mixed (abt. one hour at room temperature) or using nitrogen purge (stainless steel pipe which has a closed end and
has been bent roundthrough 180o and perforated) for a shorter (15 minutes) time.
Care is taken to mix also the very bottom layer of the barrel.
Large (> 1 m3) containers can be homogenised with a top-mounted slow revolving propeller stirrer. If circulation is used, the pump (vane pump, eccentric
pumps e.g. ‘mono’-pumps, low speed centrifugal pumps) is connected by hoses
to the bottom and upper opening of the container. One sample valve is connected
to the bottom opening. The liquid sample should be at room temperature. If the
viscosity of the liquid is still too high for pumping, a warm water (30–40°C)
circulation on the container discharge may be used. The pyrolysis liquid is
pumped from the bottom of the container and circulated for about one hour. The
homogeneity of the liquid is ensured by analysing the water and solids content
from upper and lower part of the liquid. If the difference in water and solids
between the upper and lower parts of the liquid is more than 5%, homogenisation of the sample is continued. For very viscous liquids, a propeller mixer is
recommended.
B1
B.2 Sampling and homogeneity verification
Material suitable for sample bottles/containers:
PP (polypropylene), HDPE (high-density polyethylene), PTFE (polytetrafluoroethylene) or other resistant polymeric materials, stainless steel AISI316SS, glass
(if trace levels of alkalis are not to be analysed).
Sampling
See B.1 Homogenisation. Samples are suggested to be taken at following depths:
5 vol% from surface, middle, 5 vol% above the bottom. Several sampling
devices can be used, like piston pump or large syringe. An easy and exact
method for sampling is to adjust thin glass pipes to different depths in the
sample and suck the sample slowly to a sample bottle using a slow-speed suction
pump after the sample bottle.
Microscopic determination
Samples from various layers (surface, 5 vol% from surface, middle, 5 vol% above
the bottom, bottom) of pyrolysis liquid are taken and the homogeneity is checked
by microscopy (see B.3–B.4). The very surface and bottom may be little different
from the rest of the liquid, which is accepted.
7-day standing test
The homogenised liquid sample is placed in a 100 ml measuring bottle, sealed,
and left to stand at room temperature for seven days.
The water content is analysed by Karl-Fischer titration (see Appendix C) from top,
middle, and bottom layers.
If the difference in water from top and bottom is less than 5% the sample is accepted. See B.3.
If the difference in water from top and bottom is more than 5% the sample is
abandoned. See B.4.
B2
B.3 Homogeneity determination
Good-quality liquid
The liquid is single-phase liquid determined by microscopy from various layers
(top-middle-bottom). The difference in water content by 7-day standing test is
below 5 wt% (as a% of the absolute values).
The 7-day standing test is continued. Before using the liquid batch the homogeneity of the test sample is verified.
A homogenous pyrolysis liquid by Leica DM LS microscopy.
Water, wt%
Before mixing
After mixing containers
Top
20.9
21.1
Middle
21.1
Bottom
20.7
B3
B.4 Homogeneity determination
Inhomogenous liquid
The difference in water contents in the 7-day standing test is more than 5% and
phase-separation is observed by microscopy. The liquid cannot be stored.
Poor-quality liquid
A clear phase-separation can be observed visually and liquid is abandoned.
A phase-separation can be observed by microscopy. Samples from top, middle,
and bottom layers of the unhomogenised pyrolysis liquid are taken. Water content by Karl-Fischer titration is determined from the samples and microscopic
observation is carried out. If the difference in water is more than 5% and phaseseparation is observed by microscopy the liquid is abandoned.
A phase-separated pyrolysis liquid by Leica DM LS microscopy.
aqueous phase
liquid phase
Before mixing
Top
31.1
Middle
32.4
Bottom
20.3
Water, wt%
After mixing containers
28.3
B4
B.5 Homogenisation of phase-separated pyrolysis liquid
by solvent addition (wt%). MeOH = Methanol. PEG400 =
Polyethylene-glycol 400
Addition of
5% MeOH
Phase-separated
bio-oil
Addition of
5% PEG400
Addition of 5%
Acetic Acid
B5
Appendix C:
Determination of water content
C.1 Method
Karl-Fischer titration ASTM E 203
Water addition test
Water is added to pyrolysis liquid. Samples are shaken in a mixer for about 30
minutes and left to stand overnight in air-tight bottles at room temperature. Before
water determination the samples are mixed again by hand. The water content
obtained mathematically or analytically should be the same for single-phase
liquids.
Water addition test results.
No.
0
1
2
3
4
5
6
7
Forest residue
g
10.294
10.645
10.866
11.194
11.545
11.865
12.275
Water
g
Added
0
0.317
0.588
0.869
1.183
1.547
1.873
2.272
Total
Water
wt%
Calculated
2.816
3.173
3.507
3.901
4.350
4.753
5.252
26.5
28.2
29.9
31.5
33.2
34.6
36.1
Difference
wt%
Analysed
24.3
26.4
28.2
29.6
31.3
34.0
37.0
two phases
Forest residue = VTT PDU 10/01, bottom (90 wt% of total liquid)
C1
-0.14
-0.07
-0.30
-0.19
0.77
2.35
Problem solving in Karl-Fischer titration
Observation Possible reason
Action
Fading
titration
end-point/
End-point
not clear
Change/dry the material
Check gaskets, septum
Repeat homogenisation
Use proper reagent e.g. Hydranal solvents
Wash the vessel and titrate the
water
Wet drying agent
Air leakage
Unhomogenous sample
Reactions of aldehydes/ketones
with the KF reagent
Moisture on the walls or on the
cover
Long titration Too large sample size/too much Take smaller sample
time
water
Sample dissolves slowly in KF Add dissolving solvent or exworking medium
tract the moisture from the
sample with suitable solvent
beforehand
Unclear
titration
end-point
Dirty electrode
Clean the electrode
Unhomogenous sample
Dirty solvent
Repeat homogenisation
Change the solvent, 2 - 4 determinations/solvent
Too high/low Unadequate sample homogenisawater content tion/sampling
Burette reading error
Dirty electrode
Variation in water equivalent
Wrong titration end-point
Repeat homogenisation/sampling
Calibrate burette
Clean the electrode
Check water equivalent
Calibration/water addition test
Difference in Water in titration solvent or in
duplicates
sample vessel/cover
Non-homogenous sample
Titrate the solvent
Use as large sample size as
possible
Dirty solvent
Change the solvent, 2–4 determinations/solvent
Splashing of sample to the walls Carefull sample injection
of the titration vessel
C2
Testing of sample solvent for KF water determination of forest
residue liquids
Bottom phase
Chloroform:Methanol
3:1
28.97
29.03
28.98
28.91
29.01
28.96
28.97
29.07
1:1
28.39
28.67
28.64
28.57
28.6
28.58
28.75
28.74
Average, wt%
Stdev
29.0
0.05
28.6
0.11
1:3
28.86
28.90
29.21
28.77
28.82
28.86
28.92
28.67
28.86
28.9
0.15
Top phase
Chloroform:Methanol
Average, wt%
Stdev
C3
1:3
7.80
7.73
7.75
7.76
7.80
7.75
7.71
7.70
7.8
0.04
0:1
7.81
7.82
7.86
7.73
7.88
7.79
7.82
7.73
7.8
0.05
0:1
29.03
28.96
28.76
28.98
28.79
28.98
28.86
28.90
28.9
0.10
Appendix D:
Determination of solids content
Solvent choice
•
•
Ethanol - white wood liquids from softwood or hardwood, straw pyrolysis
liquids
Methanol (CH3OH) - Dichloromethane (CH2Cl2) - extractive-rich liquids
like forest residue and bark.
Method
1. The sample size (1, 5, 10 g) is determined in order to obtain 10–20 mg dry
solid residue.
2. A representative sample of pyrolysis liquid is dissolved in about 100 ml of
solvent.
3. The solution is filtrated through a 1 µm pore size filter (Scleicher & Schöll,
GF50, ф 47 mm, Glass Fibre papers, Ref.No. 10428026, Lot: CF0401-1). The
rough side of the filter paper is faced upwards. The filter paper is soaked on to
the filter by the solvent used. If the filtration time is very long, due to high
solids content, a larger pre-filter (3 µm) may be used. The filtrate is then
filtered through a 1 µm filter. The two solids contents are combined.
4. The sample bottle, filter and residue are washed with solvent until the filtrate
is clear.
5. The filter paper with the residue is air-dried for 15 min and in an oven at
105°C for 30 minutes, cooled in a desiccator and weighed.
6. The solids content is calculated based on the original pyrolysis liquid sample.
D1
Filtration apparatus at VTT
Millipore filtration system. The original Millipore filter was too slow and was
replaced by a larger porosity metal sinter.
filter paper
metal filter
Millipore filtration system.
D2
6-place filtration system. Scleicher & Schuell, AS 600/2. Filtration system
complete with six 250 ml glass funnels with rubber lid, for 47 and 50 mm filters,
Teflon-coated support screen. A vacuum pump is used instead of water suction.
Six-place filtration system.
D3
Testing of solids content of forest residue liquids
Bottom phase
Sample
Solvent
bottom phase Methanol (MeOH)
"
Ethanol (EtOH)
"
MeOH-Dichloromethane
(MeCl2);(1:1)
"
MeOH-20 vol-% MeCl2
"
EtOH-20 vol-% MeCl2
"
EtOH-MeCl2 (1:1)
"
Isopropylalcohol (IPA)
-Industol* (1:1)
IPA-20 vol-% MeCl2
IPA-hexane (1:1)
"
"
"
"
"
"
"
"
"
"
"
Methanol20 vol-% hexane
Methanol-hexane (2:1)
Acetone
Dioxane (1,4-)-Acetone
(1:1)
Dioxane-20 vol-% IPA
Industol*
Industol20 vol-% Dioxane
Methoxy propanol
Methoxy propanol20 vol-% Dioxane
Sample size
g
15.3405
14.8577
16.1387
Residue
mg
17.70
30.20
16.30
Solids
wt-%
0.12
0.20
0.10
Average
wt-%
0.12
0.20
0.09
10.9274
10.6254
11.5852
15.9757
29.5361
30.1381
30.5504
30.0461
9.7542
9.7786
13.0855
11.1632
11.9167
10.7893
15.1620
9.30
9.40
9.30
6.20
10.30
11.20
10.90
11.20
18.10
15.30
18.70
15.70
18.80
16.50
70.70
0.09
0.09
0.08
0.04
0.03
0.04
0.04
0.04
0.19
0.16
0.14
0.14
0.16
0.15
0.47
15.5704
59.10
0.38
14.9014
16.50
0.38
Not
soluble
0.11
16.5295
15.2881
15.0626
15.50
144.20
63.30
0.09
0.94
0.42
0.09
0.94
0.42
15.1776
15.9160
15.7802
62.00
30.00
25.50
0.41
0.19
0.16
0.41
0.19
0.16
15.9512
15.8626
40.60
36.50
0.25
0.23
0.25
0.23
0.04
0.16
0.15
0.47
0.11
* Industol = Industol PE2, a trade name, 88.3% ethanol, 2.7% methyl-isobutylketone,
1.8% acetone
D4
Testing of solids content of forest residue liquids
Whole Liquid *
*Samples from the process before phase-separation, efficient mixing
before fast sampling
Sample
whole liquid 1
"
"
"
"
Solvent
Methanol
MeOH-MeCl2 (1:1)
MeOH-20 vol-% MeCl2
MeOH-hexane (2:1)
Industol20 vol-% Dioxane
"
IPA-MeCl2 (1:1)
"
Methoxy
propanol-Dioxane (1:1)
"
Industol-Dioxane (1:1)
whole liquid 2 EtOH-20 vol-% MeCl2
"
"
MeOH-MeCl2 (1:1)
"
"
"
MeOH-20 vol-% MeCl2
"
"
"
"
"
"
MeOH-10 vol-% MeCl2
"
"
EtOH
"
"
EtOH-MeCl2 (1:1)
"
"
whole liquid 3 EtOH-MeCl2 (1:1)
"
Sample size
g
10.6482
11.0005
10.9994
11.3703
10.3156
Residue
Mg
39.5
23.0
23.4
45.4
40.0
Solids
wt-%
0.37
0.21
0.21
0.40
0.39
9.4711
9.8952
43.2
36.7
0.46
0.37
0.46
0.37
9.4445
4.6223
6.1577
7.2408
4.9475
4.6785
7.0694
5.6691
6.1134
2.9725
3.1904
3.7888
5.5636
5.3646
5.0351
5.6706
5.4433
6.0428
3.1736
3.1772
3.5979
29.2
30.2
37.6
21.5
10.9
11.4
30.1
20.6
23.4
10.2
10.9
13.5
25.3
23.6
36.0
39.1
21.5
23.5
14.8
13.6
15.5
0.31
0.65
0.61
0.30
0.22
0.24
0.43
0.36
0.38
0.34
0.34
0.36
0.45
0.44
0.71
0.69
0.39
0.39
0.47
0.43
0.43
0.31
0.63
D5
Average
wt-%
0.37
0.21
0.21
0.40
0.39
0.25
0.37
0.45
0.70
0.42
0.43
Testing of solids content of forest residue liquids
Top/extractive-rich phase
Solvent
Top phase Methanol (MeOH)
"
"
"
"
"
MeOH-Dichloromethane
(MeCl2) (1:1)
"
"
"
"
MeOH-MeCl2 (20vol % MeCl2)
"
"
"
Isopropanol alcohol (IPA)
"
"
"
IPA-MeCl2 (1:1)
"
"
"
"
"
"
"
"
"
IPA-Dioxane (1:1)
"
"
"
IPA- Hexane (2:1)
"
"
"
Industol*-Dioxane (1:1)
"
"
"
MeCl2-Dioxane (1:1)
"
"
"
Methoxy propanol-Dioxane
"
(1:1)
"
Sample size
g
2.922
2.827
1.951
2.1171
3.095
0.95
Residue
Mg
77.8
73.8
53.3
57.3
83.5
10.6
Solids Average
wt-% wt-%
2.66
2.68
2.61
2.73
2.71
2.70
1.12
1.22
1.0222
1.3045
1.0924
1.677
1.7868
1.5454
2.255
2.57
2.041
2.306
2.945
3.183
2.12
2.47
1.912
1.1592
0.8016
1.586
1.348
1.223
1.274
1.325
1.1943
1.6666
1.66
1.489
1.601
1.419
1.801
1.302
1.868
1.563
1.254
12.6
16.6
13.9
34.3
36.8
30.7
107.0
121.8
84.2
20.8
28.7
31.2
19.0
22.7
18.4
13.4
9.0
18.6
10.4
9.2
9.6
112.2
106.0
143.3
12.6
11.3
11.7
69.2
100.5
66.6
12.5
12.1
9.3
1.23
1.27
1.27
2.05
2.06
1.99
4.75
4.74
4.13
0.90
0.97
0.98
0.90
0.92
0.96
1.16
1.12
1.17
0.77
0.75
0.75
8.47
8.88
8.60
0.76
0.76
0.73
4.88
5.58
5.12
0.67
0.77
0.74
* Industol is a Trade name for ethanol containing a small amount of acetone.
D6
2.03
4.54
1.01
0.76
8.65
0.75
5.19
0.73
Appendix E:
Particle size distribution for pyrolysis liquid
Equipment
Leica DM LS microscopy with the Leica Qwin Lite Standard photo analysis
programme.
Method description
Particles are marked, and the computer programme measures (based on length),
calculates and classifies them. If the solids content is high, overlapping of
particles may cause error to the results.
Sample
Forest residue liquid, PDU19/01 8.5.01 11:30 P1, solids content (insolubles in
methanol-dichloromethane, 1:1) 0.47 wt%
Measurement 1
Bin
1
2
3
4
5
6
7
8
9
10
Length
(µm)
1.00
1.58
1.58
2.51
2.51
3.98
3.98
6.31
6.31
10.0
10.0
15.8
15.8
25.1
25.1
39.8
39.8
63.1
63.1
100
Undersize Count:
Oversize Count:
713.00
0.00
Total
Mean
7 288.19
3.77
Count
(%)
613
299
472
305
131
77
23
8
2
1
Count
31.75
15.48
24.44
15.79
6.78
3.99
1.19
0.41
0.10
0.05
Total Count:
E1
1 931.00
Std Dev
Std Error
Max
Min
2-s Range
Median
Mode
Skewness
Kurtosis
Features
Spec. Area
Norm. Count
4.17
0.10
72.92
1.39
16.70
3.16
1.26
5.59
56.79
1 931.00
173 018.89
11 160.63
Measurement 2
Bin
1
2
3
4
5
6
7
8
9
10
Length
(µm)
1.00
1.58
1.58
2.51
2.51
3.98
3.98
6.31
6.31
10.0
10.0
15.8
15.8
25.1
25.1
39.8
39.8
63.1
63.1
100
Undersize Count:
Oversize Count:
Total
Mean
Std Dev
Std Error
Max
Min
2-s Range
Median
Mode
Skewness
Kurtosis
Features
Spec. Area
Norm. Count
458.00
0.00
Count
(%)
421
237
269
238
92
47
28
1
1
1
Count
31.54
17.75
20.15
17.83
6.89
3.52
2.10
0.07
0.07
0.07
Total Count:
5 108.33
3.83
4.10
0.11
65.28
1.39
16.40
3.16
1.26
4.95
47.21
1 335.00
173 018.89
7 715.92
E2
1 335.00
Measurement 3
Bin
1
2
3
4
5
6
7
8
9
10
Length
(µm)
1.00
1.58
1.58
2.51
2.51
3.98
3.98
6.31
6.31
10.0
10.0
15.8
15.8
25.1
25.1
39.8
39.8
63.1
63.1
100
Undersize Count:
Oversize Count:
Total
Mean
Std Dev
Std Error
Max
Min
2-s Range
Median
Mode
Skewness
Kurtosis
Features
Spec. Area
Norm. Count
Count
490
247
370
283
108
51
19
3
2
0
Count
(%)
31.15
15.70
23.52
17.99
6.87
3.24
1.21
0.19
0.13
0.00
510.00
0.00
Total Count:
5 829.86
3.71
3.84
0.10
60.42
1.39
15.35
3.16
1.26
5.32
51.15
1 573.00
173 018.89
9 091.49
E3
1 573.00
Appendix F:
Alkali metals and metals
Sample pretreatment:
Wet oxidation [Oasmaa et al. 1997]
The procedure must be carried out in a suitable fume cupboard. 20 ml of
concentrated (65%, pro analysis-grade) nitric acid is added to a homogenised
sample of pyrolysis liquid (3–5 g) in a conical flask. After the formation of
vapours (nitric oxides) has ceased (about 30 minutes) 10 ml of concentrated
(70%, pro analysis grade) perchloric acid is slowly added. The mixture is boiled
evenly for 1–1.5 hours until the volume of the mixture is below 5 ml and the
residue is clear and colourless. The solution is diluted with distilled water to
total volume of 50 ml. The solution is transferred to a plastic bottle and 5 ml of
hydrogen fluoride is added. The solution is let to stand overnight and analysed
by AAS or ICP. The minimum sample size needed is 10 g.
In samples with a high amount of silicates, silicon can precipitate as SiO2 during
the sample pretreatment. This may yield error in silicon. For accurate
determination of Si the sample should be ashed by dry combustion and a fusion
cake prepared from the ash for further analysis.
Standard solutions should contain the same acids in the same proportions as
when dissolving the pyrolysis liquid sample.
Detection limits for the metals
(wet oxidation and ICP-analysis):
Metal
Na
K
Ca
Mg
Fe
Si
Zn
Al
P
Detection limit
mg/kg
5–10
5–10
0.2–0.5
0.2–0.5
0.5–1
5–10
0.5–1
3–5
5–10
Metal
Pb
Cd
As
Co
Cu
Ni
V
Cr
Hg
F1
Detection limit
mg/kg
5–10
1–3
5–10
0.5–1
0.5–1
0.5–1
0.5–1
0.5–1
0.5–1
Appendix G:
Water extraction method for
pyrolysis liquids
1.
2.
3.
4.
5.
6.
8.
9.
400 g water is poured in a 500 ml Erlenmeyer or 500 ml Schott flask.
5 g of pyrolysis liquid (A, water content a wt%) is added to the water
stirring continuously by a propeller stirrer.
The bottle is closed and placed to a mixer at least for two hours. The
solution may stand overnight.
The solution is filtered through a Büchner funnel (blue ribbon filter paper,
filter pore size about 2 µm) into a suction flask. An intermediate flask,
e.g., Wulff flask, is used for water suction. Should the filtrate not be clear
it is filtered again through a 0.1 µm filter to remove tiny particles
disturbing the water analysis. If a weighable amount of precipitate
remains on the filter paper, it is dried and added (D) to the total amount of
solids.
The precipitate in the Erlenmayer/Schott flask and in the Büchner funnel
are rinsed with water.
The wet precipitate is dried in a heating oven ≤ 40oC until no further
weight loss (about 3 hours).
The precipitate is cooled in the desiccator and weighed immediately after
cooling (B).
If a weighable amount of precipitate remains on the walls of the sample
flask, the flask is dried in the vacuum oven ≤ 40oC for three hours, cooled
in an desiccator and weighed. The weight of this precipitate (C) is added
to the amount of the major part of the precipitate (B).
Analyses:
pyrolysis liquid - water
Substance insoluble in water (dry basis), wt% = (B+C+D)•100/[((100-a) •A)/100]
G1
Appendix H:
Stability test method
The pyrolysis liquid sample is mixed properly(see B.1) and let to stand, until the
air bubbles are removed. 90 ml of the sample is poured in 100 ml tight glass
bottles (or 45 ml in 50 ml bottles). The bottles are firmly closed and pre-weighed
before placing in a heating oven for a certain time. The bottles are re-tightened a
few times during the heating-up period. After a certain time the closed sample
bottles are cooled under tempered water, weighed, and analyses are performed.
The samples are mixed and measured for viscosity and water. The viscosity of
the liquid at 40°C is measured as kinematic viscosity by a standard method
(ASTM D 445). The water content is analysed by Karl Fischer titration
according to ASTM D 1744.
∆ Viscosity (40°C) [%] =
(ν2 - ν1)
__________
ν1
• 100
(ω2 - ω1)
∆ Water [%] = __________ • 100
ω1
ν1 = viscosity of the original sample, measured at 40°C, cSt
ν2 = viscosity of the aged sample, measured at 40°C, cSt
ω1 = water content of the original sample, wt%
ω2 = water content of the aged sample, wt%
Note 1:
The test is recommended to be used for internal comparison of liquid
stability for pyrolysis liquids from one process. The test is more
reliable if the initial viscosities of the samples tested are similar.
Note 2:
The possible difference in the weights before and after the test is an
indication of leakage and the test should be repeated if the net weight
loss is above 0.1 wt% of original weight.
Test conditions: 24 hours at 80°C (white wood liquids) or 1 week at 40°C (forest
residue liquids), optional conditions based on EU-IEA Round Robin 2001
[Bridgwater et al. 2001]: 6 or 24 hours at 80 °C, 1 week at 40°C.
H1
Effect of gas athmosphere, pressure, and mixing on
the viscosity of pyrolysis liquid in a stability test of
24 hours at 80°C
Test results using a 1 litre AE autoclave. 900 ml pine pyrolysis liquid.
Expt. Gas Pi
1.1
1.1.3
1.2
1.2.2
1.3
1.4
1.4.2
1.5
1.5.2
1.6
1.6.2
1.7
1.7.2
Air
Air
Air
Air
N2
N2
N2
N2
N2
N2
N2
N2
N2
Mix- T Time
Viscosity @40°C
ing
mPas
bar rpm °C
h
Ini- Final Change Aver.
tial
%
%
no
0
80 24
48 81
69
69
no
0
80 24
48 81
69
no 1.000 80 24
48 78
63
65
no 1.000 80 24
48 80
67
no
0
80 24
48 78
63
63
no 1.000 80 24
48 79
65
63
no 1.000 80 24
48 77
60
100 1.000 80 24
48 76
58
59
100 1.000 80 24
48 77
60
100
0
80 24
48 74
54
60
100
0
80 24
48 80
67
150 1.000 80 24
48 77
60
64
150 1.000 80 24
48 80
67
H2
Initial
20.9
20.9
20.9
20.9
20.9
20.9
20.9
20.9
20.9
20.9
20.9
20.9
20.9
Water
wt%
Final Change
%
22.3
6
22.4
7
22.4
7
22.0
5
22.5
8
22.2
6
22.1
6
22.3
7
22.4
7
22.8
9
22.8
9
22.3
7
22.3
7
Published by
Series title, number and report
code of publication
Vuorimiehentie 5, P.O.Box 2000, FIN-02044 VTT, Finland
Phone internat. +358 9 4561
Fax +358 9 456 4374
VTT Publications 450
VTT–PUBS–450
Author(s)
Oasmaa, Anja & Peacocke, Cordner
Title
A guide to physical property characterisation of
biomass-derived fast pyrolysis liquids
Abstract
This publication is a revised and updated version of VTT Publication 306: Physical
characterisation of biomass-based pyrolysis liquids, issued in 1997. The main purpose
of the on-going study is to test the applicability of standard fuel oil methods developed
for petroleum-based fuels to pyrolysis liquids. New methods have also been tested and
further developed. The methods were tested for pyrolysis liquids derived from hardwood, softwood, forest residue, and straw. Recommendations on liquid handling and
analyses are presented. In general, most of the standard methods can be used as such,
but the accuracy of the analyses can be improved by minor modifications. Homogeneity of the liquids is the most critical factor in the accurate analyses, and hence procedures for its verification are presented.
Keywords
pyrolysis liquid, biomass, fast pyrolysis, physical properties, fuel oil properties, test methods,
characterization, homogeneity, sampling, specifications
Activity unit
VTT Energy, New Energy Technologies, Biologinkuja 3–5, P.O.Box 1601, FIN–02044 VTT, Finland
ISBN
Project number
951–38–5878–2 (soft back ed.)
951–38–6365–4 (URL:http://www.vtt.fi/inf/pdf)
Date
December 2001
Language
English
Name of project
Pages
65 p. + app. 34
Price
B
Commissioned by
Tekes
Series title and ISSN
VTT Publications
1235–0621 (soft back ed.)
1455-0849 (URL:http://www.vtt.fi/inf/pdf)
Sold by
VTT Information Service
P.O.Box 2000, FIN-02044 VTT, Finland
Phone internat. +358 9 456 4404
Fax +358 9 456 4374
450
Denna publikation säljs av
This publication is available from
VTT INFORMATIONSTJÄNST
PB 2000
02044 VTT
Tel. (09) 456 4404
Fax (09) 456 4374
VTT INFORMATION SERVICE
P.O.Box 2000
FIN–02044 VTT, Finland
Phone internat. + 358 9 456 4404
Fax + 358 9 456 4374
ISBN 951–38–5878–2 (soft back ed.)
ISSN 1235–0621 (soft back ed.)
ISBN 951–38–6365–4 (URL: http://www.inf.vtt.fi/pdf/)
ISSN 1455–0849 (URL: http://www.inf.vtt.fi/pdf/)
A guide to physical property characterisation of biomass-derived fast pyrolysis liquids
Tätä julkaisua myy
VTT TIETOPALVELU
PL 2000
02044 VTT
Puh. (09) 456 4404
Faksi (09) 456 4374
VTT PUBLICATIONS 450
A guide to physical property characterisation of biomass-derived fast
pyrolysis liquids is a revised and updated version of VTT Publication 306:
Physical characterisation of biomass-based pyrolysis liquids. Application of
standard fuel oil analyses by Oasmaa et al. published in 1997. The objective
of this work was to test the standard methods developed for characterization
of petroleum fuels on biomass pyrolysis liquids and modify them if necessary
to accommodate the significant differences in properties between these two
groups of fuels. Compared to the previous publication, this work extends the
method validation tests on more biomass-derived liquids and more physical
properties. It includes the recommendations on testing procedures for twophase liquids obtained from pyrolysis of forest residues, the most likely
feedstock for producing pyrolysis liquids. A chapter on transporting and
handling pyrolysis liquids that considers the health and safety aspects of the
operations is included. The publication is mostly focused on practical issues
related to pyrolysis liquid characterization such as sampling techniques and
recipes for the tests. Summarizing, A guide to physical property
characterisation of biomass-derived fast pyrolysis liquids will serve as a
guide and a reference for the researchers and technicians working with
biomass pyrolysis liquids.
V T T
P U B L I C A T I O N S
Anja Oasmaa & Cordner Peacocke
A guide to physical property
characterisation of biomassderived fast pyrolysis liquids
V
T
T
P
u
b
l
i
c
a
V
T
T
P
u
b
l
i
c
a
V
T
T
P
u
b
l
i
c
a
V
T
T
P
u
b
l
i
c
a
V
T
T
P
u
b
l
i
c
a
V
T
T
P
u
b
l
i
c
V
T
T
P
u
b
l
i
V
T
T
P
u
b
l
V
T
T
P
u
b
l
P
t
i
o
n
s
t
i
o
n
s
t
i
o
n
s
t
i
o
n
s
t
i
o
n
s
a
t
i
o
n
s
c
a
t
i
o
n
s
i
c
a
t
i
o
n
s
i
c
a
t
i
o
n
s
TECHNICAL RESEARCH CENTRE OF FINLAND
ESPOO 2001